specifications | KitGuru https://www.kitguru.net KitGuru.net - Tech News | Hardware News | Hardware Reviews | IOS | Mobile | Gaming | Graphics Cards Tue, 12 Mar 2024 10:39:43 +0000 en-US hourly 1 https://wordpress.org/?v=6.4.3 https://www.kitguru.net/wp-content/uploads/2021/06/cropped-KITGURU-Light-Background-SQUARE2-32x32.png specifications | KitGuru https://www.kitguru.net 32 32 AMD’s new FreeSync specifications require FullHD monitors to have a 144Hz refresh rate https://www.kitguru.net/peripherals/monitors/joao-silva/amds-new-freesync-specifications-require-fullhd-monitors-to-have-a-144hz-refresh-rate/ https://www.kitguru.net/peripherals/monitors/joao-silva/amds-new-freesync-specifications-require-fullhd-monitors-to-have-a-144hz-refresh-rate/#respond Fri, 08 Mar 2024 13:15:09 +0000 https://www.kitguru.net/?p=650809 AMD introduced FreeSync technology in 2015 as an alternative to Nvidia's G-Sync. Since its inception, FreeSync technology has been updated multiple times and upgraded with additional levels, including a dedicated tier for HDR displays. Recently, it received yet another update, raising the minimum requirements for monitors to be FreeSync certified. As gaming display technology advances, …

The post AMD’s new FreeSync specifications require FullHD monitors to have a 144Hz refresh rate first appeared on KitGuru.]]>
AMD introduced FreeSync technology in 2015 as an alternative to Nvidia's G-Sync. Since its inception, FreeSync technology has been updated multiple times and upgraded with additional levels, including a dedicated tier for HDR displays. Recently, it received yet another update, raising the minimum requirements for monitors to be FreeSync certified.

As gaming display technology advances, AMD has chosen to upgrade the current FreeSync spec. AMD has restructured the naming scheme to make it more user-friendly and accessible for end-users while ensuring the specifications are revised appropriately.

When FreeSync technology was introduced, displays with a refresh rate of 120Hz were not common, but today, high refresh rate displays are everywhere. To keep up with the latest technology, AMD has upgraded the FreeSync specification to demand faster refresh rates. It's worth noting that while AMD has upgraded the FreeSync specifications, the modifications apply only to monitors and televisions. The company has not changed the FreeSync specifications for laptop screens. You can see the current requirements for a monitor to be FreeSync-certified below:

Certification Notebooks Monitors and TVs
FreeSync Max. Refresh Rate: 40-60 Hz < 3440 Horizontal resolution:
Max. Refresh Rate: ≥ 144 Hz
FreeSync Premium Max. Refresh Rate: ≥ 120 Hz < 3440 Horizontal resolution:
Max. Refresh Rate: ≥ 200 Hz

≥ 3440 Horizontal resolution:
Max. Refresh Rate: ≥ 120 Hz
FreeSync Premium Pro AMD FreeSync HDR on top of FreeSync Premium requirements. AMD FreeSync HDR on top of FreeSync Premium requirements.

The upgraded FreeSync specifications went live last year, but AMD has only now shared a statement regarding the changes. The minimum FreeSync specifications for 1080p displays with sub-3440 horizontal resolution have been changed to 144Hz, which was not previously a requirement for the base tier. For displays with a bigger resolution, such as 4K panels, the maximum refresh rate supported must be equal to or greater than 120Hz. Displays with sub-3440 horizontal resolution must now deliver at least 200Hz for FreeSync Premium, while Premium Pro will need to meet those requirements while also offering HDR.

Discuss on our Facebook page, HERE.

KitGuru says: Do you agree with the changes AMD made to the FreeSync spec?

The post AMD’s new FreeSync specifications require FullHD monitors to have a 144Hz refresh rate first appeared on KitGuru.]]>
https://www.kitguru.net/peripherals/monitors/joao-silva/amds-new-freesync-specifications-require-fullhd-monitors-to-have-a-144hz-refresh-rate/feed/ 0
Intel Lunar Lake processor spotted with eight cores https://www.kitguru.net/components/cpu/joao-silva/intel-lunar-lake-processor-spotted-with-eight-cores/ https://www.kitguru.net/components/cpu/joao-silva/intel-lunar-lake-processor-spotted-with-eight-cores/#respond Mon, 19 Feb 2024 16:00:09 +0000 https://www.kitguru.net/?p=648618 It would seem that Intel's upcoming Lunar Lake processors are now in testing. A recently leaked image gave us our first look at a Lunar Lake SKU, sporting eight cores. The image also seems to confirm a recent rumour that Lunar Lake will not support Hyper-Threading.  According to the screenshot shared by HXL, Lunar Lake …

The post Intel Lunar Lake processor spotted with eight cores first appeared on KitGuru.]]>
It would seem that Intel's upcoming Lunar Lake processors are now in testing. A recently leaked image gave us our first look at a Lunar Lake SKU, sporting eight cores. The image also seems to confirm a recent rumour that Lunar Lake will not support Hyper-Threading. 

According to the screenshot shared by HXL, Lunar Lake will have four Lion Cove cores and a cluster of four Skymont cores, which are also planned for usage in the Arrow Lake series. The L1 cache will have a total of 836KB. The L2 has a total of 14MB, which includes 2.5MB for each P-core and 4MB for the E-core cluster (four cores per cluster). Moreover, there are 12MB of L3 cache shared across all cores. Lastly, the screenshot suggests there's no hyper-threading.

Additionally, the leaked information reveals that the Intel Lunar Lake processor will have a base frequency of 1.8GHz and a boost clock of 2.78GHz, as seen by the 33% CPU utilisation in the screenshot. However, it may not reflect the entire clock speed, particularly as it is an early A1 silicon.

Past leaks suggested that Lunar Lake processors would come in 8W and 30W TDP versions with support for LPDDR5X-8533 memory. The most notable update on the graphics side is the upgrade to 8 Xe2-Cores, a low-power variant of the discrete Battlemage architecture.

Discuss on our Facebook page, HERE.

KitGuru says: Do you think the lack of Hyper-Threading on Intel's upcoming chips will reduce multi-core performance?

The post Intel Lunar Lake processor spotted with eight cores first appeared on KitGuru.]]>
https://www.kitguru.net/components/cpu/joao-silva/intel-lunar-lake-processor-spotted-with-eight-cores/feed/ 0
Crucial gearing up to launch its fastest SSD yet https://www.kitguru.net/components/motherboard/joao-silva/crucial-t705-pcie-5-0-ssd-to-offer-up-to-14-5-gb-s-read-speeds/ https://www.kitguru.net/components/motherboard/joao-silva/crucial-t705-pcie-5-0-ssd-to-offer-up-to-14-5-gb-s-read-speeds/#respond Tue, 06 Feb 2024 11:45:48 +0000 https://www.kitguru.net/?p=647227 Crucial looks to be cooking up its fastest PCIe 5.0 SSD yet, with a new Crucial T705 appearing online, complete with a full spec sheet. The SSD will offer up to 14.5 GB/s read speeds and will come available with a black or white heatsink.  The previous model, the T700, was released last year and …

The post Crucial gearing up to launch its fastest SSD yet first appeared on KitGuru.]]>
Crucial looks to be cooking up its fastest PCIe 5.0 SSD yet, with a new Crucial T705 appearing online, complete with a full spec sheet. The SSD will offer up to 14.5 GB/s read speeds and will come available with a black or white heatsink. 

The previous model, the T700, was released last year and was Crucial's fastest SSD when it launched. It was highly regarded as one of the best M.2 SSDs and received numerous awards, as it was mostly unrivalled by PCIe 4.0 SSDs and because other manufacturers were yet to release their first PCIe 5.0 SSDs. Even so, after other PCIe 5.0 storage drives were released, the T700 maintained its position as a top-performing SSD. However, early indications suggest that the new Crucial T705 and T705 Limited Edition will surpass its predecessor in performance.

Image credit: @momomo_us and Deepblue

Although the datasheet shared by Deepblue for the Crucial T705 does not provide details about the controller, DRAM cache, and endurance, it reveals that the SSD will be available with or without a heatsink and with 1TB, 2TB, or 4TB of storage. The Limited Edition model offers 2TB storage and comes with a white heatsink, as shown by @momomo_us.

The most significant change in the T705 spec sheet is the performance improvement, with data read rates increasing from 12,400 MB/s to 14,500 MB/s, depending on the storage capacity. The sequential write speeds were also improved, with the fastest variants of the T705 going up to 12,700 MB/s. This boost in performance is most likely due to the new controller, which has not been disclosed yet. The cache specs may also contribute to the higher performance, but we can't be sure since the details are unknown.

Discuss on our Facebook page, HERE.

KitGuru says: Do you already own a PCIe 5.0 SSD? Would you consider getting something like this Crucial T705? 

The post Crucial gearing up to launch its fastest SSD yet first appeared on KitGuru.]]>
https://www.kitguru.net/components/motherboard/joao-silva/crucial-t705-pcie-5-0-ssd-to-offer-up-to-14-5-gb-s-read-speeds/feed/ 0
Leaked slides share more details on Intel Arrow Lake-S and 800-series chipsets https://www.kitguru.net/components/cpu/joao-silva/leaked-slides-share-more-details-on-intel-arrow-lake-s-and-800-series-chipsets/ https://www.kitguru.net/components/cpu/joao-silva/leaked-slides-share-more-details-on-intel-arrow-lake-s-and-800-series-chipsets/#respond Mon, 22 Jan 2024 16:00:27 +0000 https://www.kitguru.net/?p=644494 Some of Intel's confidential slides have been leaked on social media, giving us the first glimpse of what the upcoming Arrow Lake-S and 800-series chipsets will offer. The information provided in the documents shared by YuuKi_AnS (via VideoCardz) mentions the MTL-S (Meteor Lake-S), which has been officially discontinued. However, the upcoming ARL-S (Arrow Lake-S) is …

The post Leaked slides share more details on Intel Arrow Lake-S and 800-series chipsets first appeared on KitGuru.]]>
Some of Intel's confidential slides have been leaked on social media, giving us the first glimpse of what the upcoming Arrow Lake-S and 800-series chipsets will offer.

The information provided in the documents shared by YuuKi_AnS (via VideoCardz) mentions the MTL-S (Meteor Lake-S), which has been officially discontinued. However, the upcoming ARL-S (Arrow Lake-S) is expected to use the same MTL-S baseboard. While some details in the leaked slides may be outdated, the essential aspects remain relevant.

Image credit: YuuKi_AnS

The slides only mention an 8+16+1 configuration, suggesting that Meteor Lake-S would have a maximum of 24x cores (8x P-cores and 16x E-cores). Moreover, the same slide confirms a 125 W TDP for the alleged K-series CPUs.

According to the slides, the first motherboards with the LGA-1851 socket will support DDR5-6400 memory, with no mention of DDR4 being supported. Intel's first significant change to a new platform since Alder Lake will also feature PCIe 5.0 support, with the CPU adding four more lanes for storage. The CPU will have 16 dedicated PCIe 5.0 lanes for graphics, four PCIe 5.0 lanes for storage, and four PCIe 4.0 lanes for additional storage. As for the 800-series chipset, it'll feature a total of 24 PCIe 4.0 lanes, with at least eight lanes for two M.2 drives. The diagram also confirms the presence of 4+4+2 Gen4 channels for other PCIe-connected devices, such as graphics or expansion cards. Additionally, we have confirmation of Bluetooth 6, DisplayPort 2.0 UHBR20, HDMI 2.1, USB 4, and Thunderbolt 4 support.

Intel has not made any official statements on the longevity of the LGA-1851 platform. However, the new platform is expected to remain relevant until 2026, potentially including the Panther Lake-S series.

KitGuru says: Do you think the upcoming Arrow Lake-S platform will be on par with what AMD is preparing? Are you expecting a win for the blue or the red team with the next generation of mainstream desktop CPUs?

The post Leaked slides share more details on Intel Arrow Lake-S and 800-series chipsets first appeared on KitGuru.]]>
https://www.kitguru.net/components/cpu/joao-silva/leaked-slides-share-more-details-on-intel-arrow-lake-s-and-800-series-chipsets/feed/ 0
Leaked RTX 3050 6GB specs suggest a considerable cut in performance https://www.kitguru.net/components/graphic-cards/joao-silva/leaked-rtx-3050-6gb-specs-suggest-a-considerable-cut-in-performance/ https://www.kitguru.net/components/graphic-cards/joao-silva/leaked-rtx-3050-6gb-specs-suggest-a-considerable-cut-in-performance/#respond Tue, 16 Jan 2024 10:30:33 +0000 https://www.kitguru.net/?p=643926 The rumour mill has been spinning for a while around Nvidia's reportedRTX 3050 6GB. Unlike the RTX 3050 8GB, this new model will have fewer cores, smaller memory sizes and bandwidth, and require less power. As such, it looks like the only similarity between the RTX 3050 released in 2022 and the new model is …

The post Leaked RTX 3050 6GB specs suggest a considerable cut in performance first appeared on KitGuru.]]>
The rumour mill has been spinning for a while around Nvidia's reportedRTX 3050 6GB. Unlike the RTX 3050 8GB, this new model will have fewer cores, smaller memory sizes and bandwidth, and require less power. As such, it looks like the only similarity between the RTX 3050 released in 2022 and the new model is just the name, as the performance of the 6GB variant looks to be considerably inferior.

According to VideoCardz, the upcoming RTX 3050 6GB graphics card will offer 2304 CUDA cores, which is the same as the OEM version of the RTX 3050 (the standard model has 2,560 CUDA cores). However, it has a 96-bit memory bus instead of the 128-bit bus of the other models. The corresponding reduction in memory bandwidth will drop the maximum memory bandwidth from 224 GB/s to 168 GB/s.

Moreover, the RTX 3050 6GB is expected to arrive with a 70W TDP, significantly lower than the 130W TDP of the original RTX 3050. This power reduction significantly impacts GPU clocks, with the base frequency of the new card set at 1042 MHz, a drop of 510 MHz from its predecessor. The boost clock has also been reduced considerably, but not as much, with the new card clocking in at 1470MHz, though that is still 307Hz lower than the original.

It is also worth noting that some versions of the RTX 30506 GB cards may arrive with passive cooling, which is only achievable due to the reduced power requirements of the card. The release date is expected to be within the next few weeks.

Discuss on our Facebook page HERE.

KitGuru says: After the release of the RTX 3060 6 GB variant, buyers interested in an RTX 3050 GPU should be careful about what card they're picking up. If they don't, they might come home with an unpleasant surprise.

The post Leaked RTX 3050 6GB specs suggest a considerable cut in performance first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/leaked-rtx-3050-6gb-specs-suggest-a-considerable-cut-in-performance/feed/ 0
AMD Ryzen 8000G specs leak alongside mentions of new Ryzen 5000GT CPUs https://www.kitguru.net/tech-news/featured-tech-news/joao-silva/amd-ryzen-8000g-specs-leak-alongside-mentions-of-new-ryzen-5000gt-cpus/ https://www.kitguru.net/tech-news/featured-tech-news/joao-silva/amd-ryzen-8000g-specs-leak-alongside-mentions-of-new-ryzen-5000gt-cpus/#respond Tue, 28 Nov 2023 12:00:00 +0000 https://www.kitguru.net/?p=639046 AMD looks to be preparing a new range of desktop APUs for launch. The new Ryzen 8000G series is expected to use Phoenix 1 and Phoenix 2 designs, driven by the Zen 4 and RDNA 3 architectures for CPU and GPU cores. Additionally, the first mentions of a new Ryzen 5000GT family of processors have …

The post AMD Ryzen 8000G specs leak alongside mentions of new Ryzen 5000GT CPUs first appeared on KitGuru.]]>
AMD looks to be preparing a new range of desktop APUs for launch. The new Ryzen 8000G series is expected to use Phoenix 1 and Phoenix 2 designs, driven by the Zen 4 and RDNA 3 architectures for CPU and GPU cores. Additionally, the first mentions of a new Ryzen 5000GT family of processors have been spotted. 

According to Sakhtafzarmag (via Wccftech), AMD is preparing to launch 14 new APUs based on Phoenix, including the 65W G-Series, 35W GE series, and their PRO versions. The Phoenix desktop chips will include a full Phoenix 1 die with 8 or 16 threads for Ryzen 7 8700G and Ryzen 7 PRO 8700G SKUs and a cut-down version with 6x cores and 12x threads for the Ryzen 5 8600G and its PRO variant. The 8700G is expected to have a base clock of 4.2GHz, boosting up to 5.1GHz. Compared to the 5000G series, we expect a 400 to 500MHz increase in the boost frequency across the whole lineup.

AMD is also rumoured to launch a Phoenix 2 chip (“Small Phoenix”), which will power the Ryzen 5 8500G and Ryzen 3 8300G APUs. While the specific configuration is not provided, these SKUs should include six-core (2x Zen 4 + 4x Zen4c) and quad-core (1x Zen4 + 3x Zen4c) configurations, respectively. The SKUs are expected to have boost frequencies up to 5GHz, but only for the Zen 4 cores.

In addition to the new APUs, AMD has confirmed that the Ryzen 7 5700X3D will be released on January 31st. According to rumours, the first three AMD APUs, including the 8700G, 8600G, and 8500G, will also be introduced next month, with the 8300G coming soon after for OEMs.

Finally, mentions of a new Ryzen 5000GT family have surfaced. The Ryzen 5 5600GT and Ryzen 5 5500GT are expected to be released in Q1 2024, replacing the Ryzen 5 5600G. Lastly, there are mentions of a Ryzen 5 5700 “NPU” SKU coming out by the end of January, but it's unclear how this relates to the Ryzen lineup.

Discuss on our Facebook page, HERE.

KitGuru says: Of all the upcoming AMD launches, which one interests you the most?

The post AMD Ryzen 8000G specs leak alongside mentions of new Ryzen 5000GT CPUs first appeared on KitGuru.]]>
https://www.kitguru.net/tech-news/featured-tech-news/joao-silva/amd-ryzen-8000g-specs-leak-alongside-mentions-of-new-ryzen-5000gt-cpus/feed/ 0
Intel Lunar Lake MX leak points to 8 CPU cores and a “Battlemage” iGPU https://www.kitguru.net/components/cpu/joao-silva/intel-lunar-lake-mx-leak-points-to-8-cpu-cores-and-a-battlemage-igpu/ https://www.kitguru.net/components/cpu/joao-silva/intel-lunar-lake-mx-leak-points-to-8-cpu-cores-and-a-battlemage-igpu/#respond Tue, 21 Nov 2023 09:30:14 +0000 https://www.kitguru.net/?p=638226 We're still some ways off from Intel launching its Lunar Lake chips but recent leaks give us an early idea of specifications and other upgrades. In leaked slides, it is claimed that Lunar Lake MX processors will feature TDPs ranging from 8W at the low end to 30W at most. These CPUs may also pack …

The post Intel Lunar Lake MX leak points to 8 CPU cores and a “Battlemage” iGPU first appeared on KitGuru.]]>
We're still some ways off from Intel launching its Lunar Lake chips but recent leaks give us an early idea of specifications and other upgrades. In leaked slides, it is claimed that Lunar Lake MX processors will feature TDPs ranging from 8W at the low end to 30W at most. These CPUs may also pack Intel's 2nd generation Xe GPU architecture, known as Battlemage. 

The slides were swiftly taken down from Twitter after @yuuki_ans published them. However, an Anandtech forum user was fast enough to copy them all before they were removed. The new CPU series, with a TDP ranging between 8W and 30W, is intended for low-power devices and should replace Meteor Lake-U, which has not formally launched yet.

According to Intel, Lunar Lake was co-developed with Microsoft to create a more power-efficient software/hardware integration. The new architecture supports dual-channel LPDDR5X-8533 memory and an on-package memory (MoP). Intel has integrated the memory, reducing the size (about 100–250 mm2) and power consumption (reduced routing, integrated memory subsystem). The package will include two tiles: one for the CPU and GPU and another for the SoC. Its dimensions are 27.5mm by 27mm.

Based on the slides, Lunar Lake's integrated memory will vary between 16GB and 32GB. The CPU architecture will combine Skymont and Lion Cove cores and could be based on TSMC N3B process technology. Intel intends to release at least four Lunar Lake SKUs:

Series Core Configuration Memory GPU configuration Cache NPU configuration
Core 5 4P + 4E 16 GB 7 Xe LPG Cores 8 MB 5-tile NPU
Core 5 4P + 4E 32 GB 7 Xe LPG Cores 8 MB 5-tile NPU
Core 7 4P + 4E 16 GB 8 Xe LPG Cores 8 MB 6-tile NPU
Core 7 4P + 4E 32 GB 8 Xe LPG Cores 8 MB 6-tile NPU

The new CPUs will feature an advanced built-in NPU 4.0 with Xe2-LPG (Battlemage) integrated graphics. As such, the Lunar Lake MX series should be the first to launch a power-efficient version of the “Battlemage” GPU, with up to 8 Xe-Cores included in the Core 7 series. This GPU architecture will support real-time ray tracing, Systolic AI/Super Scaling technologies, and VVC/H.266 video decoding. Moreover, it will have 4x PCIe 5.0 lanes and 4x PCIe 4.0 lanes. The Lunar Lake MX series will also support DisplayPort 1.4, HDMI 2.1, eDP1.4 and 1.5, Thunderbolt 4, and up to three USB4 ports. A BE201 network card with a CNVio3 interface supports Bluetooth 5.4 and WiFi-7.

Interestingly, the Lunar Lake 8W SKUs can run without a fan, while the fanned versions will run with a TDP between 17 and 30W. Moreover, the 12W version of the Arc Xe2-LPG GPU is capable of 2.5 TFLOPS of single-precision computation, about the same as an Apple M1. However, this chip's maximum Arc performance may reach 3.8 TFLOPS, enough to outperform the Apple M2. The Lunar Lake MX series is expected to be integrated into next-generation portable devices in 2024 or 2025.

Discuss on our Facebook page, HERE.

KitGuru says: Lunar Lake is shaping up to be a decent alternative to some Apple M devices. However, it seems it won't be enough to perform at the same level as the M3. That shouldn't be an issue if manufacturers don't try to sell their devices at the same price as an M3-powered MacBook.

The post Intel Lunar Lake MX leak points to 8 CPU cores and a “Battlemage” iGPU first appeared on KitGuru.]]>
https://www.kitguru.net/components/cpu/joao-silva/intel-lunar-lake-mx-leak-points-to-8-cpu-cores-and-a-battlemage-igpu/feed/ 0
AMD is reportedly launching Ryzen 7 5700X3D and Ryzen 5 5500X3D CPUs https://www.kitguru.net/components/cpu/joao-silva/amd-is-reportedly-launching-ryzen-7-5700x3d-and-ryzen-5-5500x3d-cpus/ https://www.kitguru.net/components/cpu/joao-silva/amd-is-reportedly-launching-ryzen-7-5700x3d-and-ryzen-5-5500x3d-cpus/#respond Wed, 15 Nov 2023 09:30:21 +0000 https://www.kitguru.net/?p=637576 It looks like the AMD AM4 platform hasn't been forgotten yet. As per the latest rumours, AMD plans to launch two new Ryzen 5000X3D CPUs for budget-conscious gamers, making the mid-range and entry-level market segments more attractive for those looking to build a new PC. According to chi11eddog, AMD is reportedly working on two new …

The post AMD is reportedly launching Ryzen 7 5700X3D and Ryzen 5 5500X3D CPUs first appeared on KitGuru.]]>
It looks like the AMD AM4 platform hasn't been forgotten yet. As per the latest rumours, AMD plans to launch two new Ryzen 5000X3D CPUs for budget-conscious gamers, making the mid-range and entry-level market segments more attractive for those looking to build a new PC.

According to chi11eddog, AMD is reportedly working on two new Ryzen 5000X3D series CPUs. One is the Ryzen 7 5700X3D, an affordable 8-core variant with a base frequency of 3.0GHz and a peak clock of 4.1GHz, 400 MHz lower than the 5800X3D SKU. The second rumoured CPU is the Ryzen 5 5500X3D, a 6-core model with base and boost frequencies of 3.0 and 4.0GHz, respectively. These clock rates are believed to be 300/400 MHz slower than the 5600X3D. Both of these CPUs are said to feature 96MB of 3D V-Cache.

It's unclear when these new CPUs will be available. Still, AMD already has two 5000 series CPUs featuring 3D V-Cache: the 5800X3D, which is considered one of the best gaming CPUs for the AM4 platform, and the 5600X3D, which had a limited launch this year. If AMD expands the availability of 3D V-Cache options beyond select shops, it would be very welcome for those looking for more CPU options in the mid and low-end segments.

What sets the X3D series apart is its compatibility with various entry-level motherboards. Since these CPUs don't allow overclocking, their system requirements are lower. Despite this limitation, X3D CPUs outperform their non-X3D counterparts in gaming benchmarks, whether overclocked or not.

Discuss on our Facebook page, HERE.

KitGuru says: How do you think these two new Ryzen 5000X3D processors will compare with the rest of the lineup? Assuming the specs are accurate, would you prefer the extra cores of the new Ryzen 7 chip over the higher clock speeds of the Ryzen 5 5600X3D?

The post AMD is reportedly launching Ryzen 7 5700X3D and Ryzen 5 5500X3D CPUs first appeared on KitGuru.]]>
https://www.kitguru.net/components/cpu/joao-silva/amd-is-reportedly-launching-ryzen-7-5700x3d-and-ryzen-5-5500x3d-cpus/feed/ 0
Leak details Intel Xeon W-2500 HEDT CPU specs https://www.kitguru.net/components/cpu/joao-silva/leak-details-intel-xeon-w-2500-hedt-cpu-specs/ https://www.kitguru.net/components/cpu/joao-silva/leak-details-intel-xeon-w-2500-hedt-cpu-specs/#respond Sun, 05 Nov 2023 11:30:23 +0000 https://www.kitguru.net/?p=636420 A new leak claims to show Intel's full line-up of upcoming W-2500 HEDT CPUs, otherwise known as the Sapphire Rapids Refresh. It looks like we'll be seeing seven new processors to start with, each with core count and cache upgrades over their predecessors.  As per the table shared by @yuuki_ans, the upcoming “Sapphire Rapids Refresh” …

The post Leak details Intel Xeon W-2500 HEDT CPU specs first appeared on KitGuru.]]>
A new leak claims to show Intel's full line-up of upcoming W-2500 HEDT CPUs, otherwise known as the Sapphire Rapids Refresh. It looks like we'll be seeing seven new processors to start with, each with core count and cache upgrades over their predecessors. 

As per the table shared by @yuuki_ans, the upcoming “Sapphire Rapids Refresh” series will be led by the Xeon W7-2595X CPU, boasting 26 cores and a boost clock of up to 4.8GHz. Compared to the previous series' flagship, there's a 2-core increase, and the L3 cache has been increased to 48.75MB. The base clock has also been boosted from 2.5 to 2.8GHz, while the highest clock remains the same as that of the W7-2495X. Notably, the TDP has been increased from 225W to 250W.

An analysis of the table provided by the leaker reveals that the new series features a straightforward trend of a higher number of cores (2 cores) and higher base clocks. Some components have also received higher boost clocks. Interestingly, all processors have a higher TDP, with an increase ranging from 10W to 25W. The CPUs are compatible with the LGA-4677 socket and can be used with motherboards that are currently in use and based on W790.

While the W3xxx series is designed for high-end workstations and content creation, the W2xxx series is primarily aimed at hobbyists and overclocking enthusiasts. However, it's unclear if Intel plans to replace its 26C+ core series, as the W3500 series is not mentioned in the leak.

Discuss on our Facebook page, HERE.

KitGuru says: Contrary to the Raptor Lake Refresh, which only got higher clock speeds and an increased core count in select models, this refresh looks more significant. Besides the higher base and boost clock speeds, all new chips also get more cores and cache.

The post Leak details Intel Xeon W-2500 HEDT CPU specs first appeared on KitGuru.]]>
https://www.kitguru.net/components/cpu/joao-silva/leak-details-intel-xeon-w-2500-hedt-cpu-specs/feed/ 0
New RTX 40 Super rumours suggest small performance uplift https://www.kitguru.net/components/graphic-cards/joao-silva/new-rtx-40-super-rumours-suggest-small-performance-uplift/ https://www.kitguru.net/components/graphic-cards/joao-silva/new-rtx-40-super-rumours-suggest-small-performance-uplift/#respond Tue, 31 Oct 2023 17:30:17 +0000 https://www.kitguru.net/?p=635615 This past week has been abuzz with rumours about the apparently upcoming RTX 40 Super graphics cards. The latest one comes from a known Nvidia leaker, who has revealed the expected specs for three RTX 40 Super models. These specs include the CUDA core specifications and probable GPU models for these cards. According to Kopite, …

The post New RTX 40 Super rumours suggest small performance uplift first appeared on KitGuru.]]>
This past week has been abuzz with rumours about the apparently upcoming RTX 40 Super graphics cards. The latest one comes from a known Nvidia leaker, who has revealed the expected specs for three RTX 40 Super models. These specs include the CUDA core specifications and probable GPU models for these cards.

According to Kopite, the RTX 4080 Super will feature a fully enabled AD103-300 GPU with 10,240 CUDA cores. While the memory configuration remains unknown, it's worth noting that the AD103 GPU indicates a 256-bit memory bus.

Moving to the weirdly named RTX 4070 Ti Super, this card is expected to pack 8,448 CUDA cores and 48MB of L2 cache within one of two GPU variants: the AD103-275 or the AD102-175. The leaker also added that this GPU will have a 256-bit memory bus. As for the RTX 4070 Super, it's expected to feature 7,168 CUDA cores. This variation, like the previously described SKU, might feature two GPU options: the AD104-350 or the AD103-175.

Apparently, these specs are several weeks old, so rather than two GPU versions, Nvidia might be still deciding between versions of its ADA GPUs before launch. It has also been reported that board partners have not been told about RTX 40 Super yet, indicating that a launch won't happen until next year, if at all.

KitGuru says: Are you interested in these new possible Nvidia RTX 40 Super graphics cards? If so, which looks the most interesting to you?

The post New RTX 40 Super rumours suggest small performance uplift first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/new-rtx-40-super-rumours-suggest-small-performance-uplift/feed/ 0
RTX 5090 rumoured be about 70% faster than the RTX 4090 https://www.kitguru.net/components/graphic-cards/joao-silva/rtx-5090-rumoured-be-about-70-faster-than-the-rtx-4090/ https://www.kitguru.net/components/graphic-cards/joao-silva/rtx-5090-rumoured-be-about-70-faster-than-the-rtx-4090/#respond Tue, 19 Sep 2023 07:27:14 +0000 https://www.kitguru.net/?p=630885 While early reports have already claimed that Nvidia will wait until 2025 to release the RTX 5090, the first rumours about the upcoming GPU are already making the rounds online - including reports that it could be as much as 70% faster than the RTX 4090.

The post RTX 5090 rumoured be about 70% faster than the RTX 4090 first appeared on KitGuru.]]>
While early reports have already claimed that Nvidia will wait until 2025 to release the RTX 5090, the first rumours about the upcoming GPU are already making the rounds online – including reports that it could be as much as 70% faster than the RTX 4090.

The rumours were first shared by Panzerlied (via VideoCardz), the same leaker who shared the first details on Nvidia's next-generation of consumer GPUs and that NVIDIA intends to skip the xxx04-class GPUs in their GPU series. Following Panzerlied's post, VideoCardz asked kopite7kimi about these rumours, which the leaker corroborated.

According to Panzerlied (based on DeepL translation), the RTX 5090 would have a 50% increase in scale, a 52% increase in memory bandwidth, a 78% increase in L2 cache, and 15% higher clock speeds. All this would reportedly net a 70% improvement in performance.

Taking the RTX 4090 specs into account, a 52% memory bandwidth increase (about 1.54 TB/s) means we could see the RTX 5090 using either 24 GB of GDDR7 memory running at 32 Gbps. Other alternatives would be using a 512-bit memory bus with 24 Gbps memory or a 448-bit memory bus with 28 Gbps memory. As for the clock speed, it's possible the upcoming flagship card will boost close to the 3.0 GHz mark, and in some cases, even surpass it. Lastly, a 78% increase in L2 cache means we can expect 128 MB of L2 cache on the RTX 5090.

Although all this looks promising, it will take time to get confirmation on all of this. After all, the RTX 5090 and other RTX 50 series GPUs are expected to be released in 2025.

Discuss on our Facebook page HERE.

KitGuru says: What kind of generational uplift are you expecting from the RTX 50 series? Does a 70% improvement over the already mighty RTX 4090 seem plausible?

The post RTX 5090 rumoured be about 70% faster than the RTX 4090 first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/rtx-5090-rumoured-be-about-70-faster-than-the-rtx-4090/feed/ 0
Screenshot shows Intel Core i5-14600K with 14 cores and 5.3GHz clock speed https://www.kitguru.net/components/cpu/joao-silva/screenshot-shows-intel-core-i5-14600k-with-14-cores-and-5-3ghz-clock-speed/ https://www.kitguru.net/components/cpu/joao-silva/screenshot-shows-intel-core-i5-14600k-with-14-cores-and-5-3ghz-clock-speed/#respond Wed, 09 Aug 2023 16:00:09 +0000 https://www.kitguru.net/?p=625805 After confirmation of key specs for the upcoming Intel Core i9-14900K and i7-14700K, it's now time to learn more about the Core i5-14600K. The latest leak gives us a look at the new-gen unlocked Core i5, with 14 CPU cores, alongside a 5.3GHz clock speed.  As per the picture shared by HXL, we can confirm …

The post Screenshot shows Intel Core i5-14600K with 14 cores and 5.3GHz clock speed first appeared on KitGuru.]]>
After confirmation of key specs for the upcoming Intel Core i9-14900K and i7-14700K, it's now time to learn more about the Core i5-14600K. The latest leak gives us a look at the new-gen unlocked Core i5, with 14 CPU cores, alongside a 5.3GHz clock speed. 

As per the picture shared by HXL, we can confirm some of the details about the upcoming CPU via the CPU-Z and Core Temp outputs. If the information shown in the image is genuine, it seems that the CPU's core count (6P + 8E) will stay the same as its predecessors, as expected. However, it's almost guaranteed that an increase in frequency is due.

The latest rumoured specs indicated that the Core i5-14600K will boost up to 5.3GHz, 200 MHz more than the Core i5-13600K. Looking at the information displayed in CPU-Z, it matches what we've been hearing regarding the CPU's boost clock speed. Note that the system that we see in the picture is using the F5 BIOS, with the CPU running on a Gigabyte Z790 Gaming X Axe motherboard.

Intel is getting ready to introduce its new desktop CPU series in September, with availability likely following in October.

KitGuru says: Are any of you planning on upgrading your PC later this year when new hardware drops? 

The post Screenshot shows Intel Core i5-14600K with 14 cores and 5.3GHz clock speed first appeared on KitGuru.]]>
https://www.kitguru.net/components/cpu/joao-silva/screenshot-shows-intel-core-i5-14600k-with-14-cores-and-5-3ghz-clock-speed/feed/ 0
ASRock listings confirm Radeon RX 7800 XT and RX 7700 XT https://www.kitguru.net/components/graphic-cards/joao-silva/asrock-listings-confirm-radeon-rx-7800-xt-and-rx-7700-xt/ https://www.kitguru.net/components/graphic-cards/joao-silva/asrock-listings-confirm-radeon-rx-7800-xt-and-rx-7700-xt/#respond Wed, 09 Aug 2023 13:00:04 +0000 https://www.kitguru.net/?p=625804 ASRock has once again submitted applications to the Eurasian Economic Commission (EEC) for the Radeon RX 7000 series. The most current information obtained from these submissions provides a peek at two Radeon desktop GPUs from AMD that have yet to be unveiled. It's important to note that ASRock has already submitted an entry to EEC …

The post ASRock listings confirm Radeon RX 7800 XT and RX 7700 XT first appeared on KitGuru.]]>
ASRock has once again submitted applications to the Eurasian Economic Commission (EEC) for the Radeon RX 7000 series. The most current information obtained from these submissions provides a peek at two Radeon desktop GPUs from AMD that have yet to be unveiled.

It's important to note that ASRock has already submitted an entry to EEC referencing RX 7800 XT GPUs. Previous postings included a handful of Phantom Gaming versions, clearly indicating that one of AMD's next cards would have a 16GB memory configuration. However, the latest listing (via @momomo_us) confirms the existence of more models that span the Steel Legend (SL) and Challenger (CL) series.

The new listing also suggests the RX 7700 XT should release relatively soon while also confirming how much VRAM it has. In line with the RX 6700 XT, the upcoming GPU is expected to have a memory capacity of 12GB. Moreover, the models listed suggest that there will be RX 7700 XT models belonging to the Phantom Gaming (PG), SL, and CL series.

It's important to remember that EEC listings don't always reflect the actual specification since companies sometimes list models that never see the light of day. It's speculated that the RX 7800 XT and RX 7700 XT will release this quarter.

KitGuru says: The RX 7000 series has seen a slow rollout but after many months, the 7700 and 7800 series GPUs are finally set to arrive. Will any of you be looking to upgrade once these launch? 

The post ASRock listings confirm Radeon RX 7800 XT and RX 7700 XT first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/asrock-listings-confirm-radeon-rx-7800-xt-and-rx-7700-xt/feed/ 0
Intel Meteor Lake iGPU will reportedly offer more compute power than the Radeon 780M https://www.kitguru.net/components/cpu/joao-silva/intel-meteor-lake-igpu-will-reportedly-offer-more-compute-power-than-the-radeon-780m/ https://www.kitguru.net/components/cpu/joao-silva/intel-meteor-lake-igpu-will-reportedly-offer-more-compute-power-than-the-radeon-780m/#respond Wed, 12 Jul 2023 15:00:24 +0000 https://www.kitguru.net/?p=622811 Intel Meteor Lake is still a few months away from coming to the market, but as the alleged release period (Q4 2023) approaches, reports about the upcoming CPUs have been increasing. Most recently, details on the Meteor Lake iGPU and CPU clock speeds have been revealed via a qualification sample.  Intel is nearing the final …

The post Intel Meteor Lake iGPU will reportedly offer more compute power than the Radeon 780M first appeared on KitGuru.]]>
Intel Meteor Lake is still a few months away from coming to the market, but as the alleged release period (Q4 2023) approaches, reports about the upcoming CPUs have been increasing. Most recently, details on the Meteor Lake iGPU and CPU clock speeds have been revealed via a qualification sample. 

Intel is nearing the final stages of preparation for the launch of its Meteor Lake chips later this year. The upcoming mobile Core Ultra series will introduce a new architecture based on Intel 4 process technology, featuring a hybrid design incorporating Redwood Cove and Crestmont cores and an iGPU based on Xe-LPG architecture.

As per Golden Pig Upgrade (via VideoCardz), the qualification sample of Meteor Lake currently operates within a power range of 20 to 65W and can achieve a boost clock of up to 4.8GHz. While this falls short of the 5.0GHz boost of the Core i7-1370P (Raptor Lake) and Intel's current offerings, the source suggests that the aim is to reach 5.0GHz with Core Ultra 9 processors.

The compute die of Meteor Lake consists of 6x P-cores, 8x E-cores, and 2x low-power E-cores. The GPU subsystem will feature 8 Xe Cores with 128 EUs (1024 shading units). The reported boost clock of the GPU is 2.2GHz, resulting in a single-precision compute power of 4.5 TFLOPS. That's 5% higher than what the Radeon 780M RDNA3 iGPU found in Phoenix APUs is capable of.

Discuss on our Facebook page, HERE.

KitGuru says: AMD Phoenix APUs have found success in small handheld gaming systems. Assuming the Meteor Lake iGPU will be as good as the report suggests, handheld system manufacturers will certainly evaluate Intel's new chips for new and upcoming portable consoles.

The post Intel Meteor Lake iGPU will reportedly offer more compute power than the Radeon 780M first appeared on KitGuru.]]>
https://www.kitguru.net/components/cpu/joao-silva/intel-meteor-lake-igpu-will-reportedly-offer-more-compute-power-than-the-radeon-780m/feed/ 0
Leaked Intel slides detail Meteor Lake mobile specs https://www.kitguru.net/components/cpu/joao-silva/leaked-intel-slides-detail-meteor-lake-mobile-specs/ https://www.kitguru.net/components/cpu/joao-silva/leaked-intel-slides-detail-meteor-lake-mobile-specs/#respond Sun, 09 Jul 2023 15:29:18 +0000 https://www.kitguru.net/?p=622452 Recently leaked slides from Intel have unveiled the core count for the upcoming Meteor Lake mobile CPU lineup. While the company had previously confirmed the use of Raptor Lake Refresh for top-tier chips, it appears that the majority of the series will rely on the all-new Meteor Lake lineup. The slides depict the various tiers …

The post Leaked Intel slides detail Meteor Lake mobile specs first appeared on KitGuru.]]>
Recently leaked slides from Intel have unveiled the core count for the upcoming Meteor Lake mobile CPU lineup. While the company had previously confirmed the use of Raptor Lake Refresh for top-tier chips, it appears that the majority of the series will rely on the all-new Meteor Lake lineup. The slides depict the various tiers of the CPU series, showcasing the continuation of six power segments.

The slides were shown in a video by Moore's Law Is Dead, which also claims details on upcoming Raptor Lake Refresh processors and next-gen Intel Arc graphics cards.

The following table details the specifications for each of the six power tiers that will comprise the next generation of Intel mobile chips:

PL1 TDP Core Configurations Total Cores iGPU Silicon Series Package
7W 1P + 4E, 1P + 8E 5 to 9 3 Xe – 4 Xe Meteor Lake MTL-U M
9W 2P+4E, 2P + 8E 6 to 10 3 Xe – 4 Xe Meteor Lake MTL-U M
15W 2P+4E, 2P + 8E, 4P + 8E 6 to 12 3 Xe – 8 Xe Meteor Lake MTL-U P
28W 2P + 8E, 4P + 8E, 6P + 8E 10 to 14 7 Xe – 8 Xe Meteor Lake MTL-P P
45W 4P + 8E, 6P + 8E 12 to 14 8 Xe Meteor Lake MTL-H P
55W 6P + 8E, 8P + 8/12E, 8P + 16E 14 to 24 32 EU Raptor Lake RPL-HX S-BGA

Interestingly, although Intel has emphasised the Meteor Lake architecture, the last column suggests Raptor Lake chips will still be part of the next-gen mobile CPU lineup. Aside from the difference in CPU core count, the notable distinction lies in the GPU upgrade for Meteor Lake, which features the advanced Xe-LPG architecture. Conversely, the higher-end SKUs with Raptor Lake will retain the architecture of the current iteration, with a limitation of 32 units.

The Xe-LPG integrated graphics will feature up to 8 Xe units, which translates into 128 Vector Engines and 1024 FP32 cores (same as the Arc A380). Additionally, a new slide introduces the Xe-LPG architecture, highlighting advancements in power and area optimisations, support for advanced DX12U graphics, and a significant 33% increase in graphics core capabilities.

Discuss on our Facebook page, HERE.

KitGuru says: Seeing as the HX chips will still be based on Raptor Lake, we wouldn't be surprised if some Meteor Lake-H chips were able to outperform them, more so in single-core workloads.

The post Leaked Intel slides detail Meteor Lake mobile specs first appeared on KitGuru.]]>
https://www.kitguru.net/components/cpu/joao-silva/leaked-intel-slides-detail-meteor-lake-mobile-specs/feed/ 0
AMD RX 7600 specifications confirmed by GPU-Z screenshot https://www.kitguru.net/components/graphic-cards/joao-silva/amd-radeon-rx-7600-specifications-confirmed-by-gpu-z-screenshot/ https://www.kitguru.net/components/graphic-cards/joao-silva/amd-radeon-rx-7600-specifications-confirmed-by-gpu-z-screenshot/#respond Fri, 19 May 2023 16:50:33 +0000 https://www.kitguru.net/?p=615420 New information about the upcoming RX 7600 graphics card has surfaced, primarily thanks to a new GPU-Z screenshot detailing the upcoming GPU. Although the software does not officially support this particular graphics card yet, it has already introduced Navi 33 support, allowing us to see the correct GPU specifications...

The post AMD RX 7600 specifications confirmed by GPU-Z screenshot first appeared on KitGuru.]]>
New information about the upcoming RX 7600 graphics card has surfaced, primarily thanks to a new GPU-Z screenshot detailing the upcoming GPU. Although the software does not officially support this particular graphics card yet, it has already introduced Navi 33 support, allowing us to see the correct GPU specifications.

According to the screenshot shared by VideoCardz, the RX 7600 will be powered by the Navi 33 XL GPU, packing 32 CUs and 2048 stream processors. The graphics card will also be equipped with 8GB of GDDR6 memory running at 18Gbps on a 128-bit memory bus. Furthermore, it has been revealed that the RX 7600 will have a limited PCIe interface, operating across eight PCIe 4.0 lanes, mirroring the setup of the RTX 4060 series.

Image credit: VideoCardz

Regarding clock speeds, the GPU-Z software provides some exciting figures. The base clock is reportedly 1,720MHz, while the game clock is 2,250MHz. If thermal requirements are met, the boost clock can reach an impressive 2,655MHz. However, the GPU clock can reportedly go up to 2,850MHz in real-world usage.

The fact that this GPU has a Navi 33 XL chip suggests that we might see a Navi 33 XT-based GPU (Radeon RX 7600 XT?). If that's the case, we can expect more CUs and, hopefully, more VRAM.

Discuss on our Facebook page, HERE.

KitGuru says: The RX 7600 will likely be going up against the newly announced RTX 4060, so there will be an interesting battle ahead in the midrange market. 

The post AMD RX 7600 specifications confirmed by GPU-Z screenshot first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/amd-radeon-rx-7600-specifications-confirmed-by-gpu-z-screenshot/feed/ 0
Nvidia RTX 4070 has been spotted on Geekbench ahead of release https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4070-has-been-spotted-on-geekbench-ahead-of-release/ https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4070-has-been-spotted-on-geekbench-ahead-of-release/#respond Wed, 05 Apr 2023 08:00:21 +0000 https://www.kitguru.net/?p=610331 With the launch date of the RTX 4070 approaching, it's not surprising to see it appear on benchmark databases. The new entry on the Geekbench website confirms some of the graphics card specs that were still up for debate, including CUDA core count, memory speed, and GPU boost clock speed. The Geekbench entry (CUDA) spotted …

The post Nvidia RTX 4070 has been spotted on Geekbench ahead of release first appeared on KitGuru.]]>
With the launch date of the RTX 4070 approaching, it's not surprising to see it appear on benchmark databases. The new entry on the Geekbench website confirms some of the graphics card specs that were still up for debate, including CUDA core count, memory speed, and GPU boost clock speed.

The Geekbench entry (CUDA) spotted by Benchleaks shows the Nvidia RTX 4070 running on an ASRock Z790 Taichi Carrara motherboard with 32GB of DDR5-6000 memory and an Intel Core i9-13900K. In addition to that entry, there's another one for the OpenCL test, but it was deemed invalid by the database “due to an issue with the timers on this system.”

Looking at the specs of the GPU on these entries, we can confirm a few of the remaining rumoured specs of the RTX 4070. For example, we already knew that the card would feature 12GB of GDDR6X memory. However, we only have confirmation that the memory will run at 21Gbps with this entry. Moreover, it confirms the 2,475MHz GPU boost clock speed and the 5888 CUDA core count (46 multiprocessors x 128 cores).

The GPU scored 202,437 points in the CUDA test, putting it on par with the RTX 3080 10G. Compared with its predecessor, the upcoming graphics card scored 25-30% higher. As for the OpenCL test score, even if it was considered invalid, it falls within the expected values. Scoring 177,594 points, it's within the margin of error of the RTX 3080 10G and 12G GPUs. When compared to the RTX 3070, the RTX 4070 scored almost 40% higher.

The Nvidia RTX 4070 is expected to be announced on April 12th and released on April 13th. The GPU's MSRP will reportedly be set at $599.

Discuss on our Facebook page, HERE.

KitGuru says: Geekbench is far from being an accurate representation of gaming performance. However, it has proven to be somewhat accurate when comparing GPUs from the same manufacturer. Taking that into consideration, getting an RTX 3080 with DLSS 3 and an extra 2GB of VRAM for $100 less doesn't seem that bad of a deal.

The post Nvidia RTX 4070 has been spotted on Geekbench ahead of release first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4070-has-been-spotted-on-geekbench-ahead-of-release/feed/ 0
Leaked slide shows Nvidia RTX 4070 specifications and pricing https://www.kitguru.net/components/graphic-cards/joao-silva/leak-confirms-nvidia-rtx-4070-specifications-and-pricing/ https://www.kitguru.net/components/graphic-cards/joao-silva/leak-confirms-nvidia-rtx-4070-specifications-and-pricing/#respond Mon, 03 Apr 2023 14:00:44 +0000 https://www.kitguru.net/?p=610054 Some marketing material for the Nvidia RTX 4070 announcement has seemingly leaked, confirming the majority of the latest rumours regarding the pricing and specifications of the GPU. Moreover, it shares a few other details, including the amount of L2 cache, the average gaming power, idle power, and video playback power. The picture shared by VideoCardz …

The post Leaked slide shows Nvidia RTX 4070 specifications and pricing first appeared on KitGuru.]]>
Some marketing material for the Nvidia RTX 4070 announcement has seemingly leaked, confirming the majority of the latest rumours regarding the pricing and specifications of the GPU. Moreover, it shares a few other details, including the amount of L2 cache, the average gaming power, idle power, and video playback power.

The picture shared by VideoCardz confirms what we've been hearing for a while and shares other specs from the upcoming mid-range graphics card. As you can see by the picture, the RTX 4070 will be based on the AD104 GPU with 400 tensor cores (4th gen) and 87 RT cores (3rd gen). CUDA core count isn't included in the slide.

Image credit: VideoCardz

Besides the 12GB of GDDR6X memory, it looks like the RTX 4070 will also pack 36MB of L2 cache, 32MB more than the RTX 3070 Ti and non-Ti. That could compensate for the lower maximum bandwidth of the upcoming model, which is limited to 504GB/s (192-bit memory bus). TGP has been confirmed to be 200W.

Compared to its predecessors, the average gaming power was reduced to 186W, down from 215W on the RTX 3070 and 240W on the RTX 3070 Ti. Video playback power was also improved to 16W, 4W less than on the RTX 3070 series GPUs. Lastly, idle power consumption got a slight decrease to 10W.

As per the leaked slide, the RTX 4070 will release with an MSRP of $599, which matches the latest reports regarding the GPU's pricing. Nvidia is said to be launching the RTX 4070 in mid April but the company has yet to make an official announcement.

WKitGuru says: Assuming all of this information is accurate, do you think the RTX 4070 will be worth getting at MSRP, or is it too high? 

The post Leaked slide shows Nvidia RTX 4070 specifications and pricing first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/leak-confirms-nvidia-rtx-4070-specifications-and-pricing/feed/ 0
Nvidia RTX 4060 will use an unlocked AD107 GPU according to leak https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4060-will-use-an-unlocked-ad107-gpu-according-to-leak/ https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4060-will-use-an-unlocked-ad107-gpu-according-to-leak/#respond Tue, 14 Feb 2023 14:32:27 +0000 https://www.kitguru.net/?p=593453 The RTX 4060 specifications have seemingly been leaked. As per the report, the upcoming graphics card will be based on Nvidia's smallest Ada GPU, the AD107 GPU. Assuming that's true, it means the desktop and laptop variants of the RTX 4060 will use the same GPU, unlike the other RTX 40 series SKUs. According to …

The post Nvidia RTX 4060 will use an unlocked AD107 GPU according to leak first appeared on KitGuru.]]>
The RTX 4060 specifications have seemingly been leaked. As per the report, the upcoming graphics card will be based on Nvidia's smallest Ada GPU, the AD107 GPU. Assuming that's true, it means the desktop and laptop variants of the RTX 4060 will use the same GPU, unlike the other RTX 40 series SKUs.

According to kopite7kimi, the desktop variant of the RTX 4060 will pack a fully enabled AD107 GPU (AD107-400-A1) with 3,072 CUDA cores and 24MB of L2 cache on a PG190 board. For the memory, there's 8GB of GDDR6 memory clocked at 18Gbps across a 128-bit memory bus, offering a maximum bandwidth of 288GB/s.

All this will be packed in a 115W TDP, the same as its laptop variant and 55W less than the RTX 3060 desktop GPU. So, even if it offers the same performance as the RTX 3060, one thing that has definitely been improved is efficiency.

We've yet to hear anything about the release date of the RTX 4060, but considering we are already getting some detailed specs of the GPU, there's a chance we might see it coming out in the first half of the year.

Discuss on our Facebook page, HERE.

KitGuru says: Considering the rumoured specs, what kind of performance are you expecting from the RTX 4060?

The post Nvidia RTX 4060 will use an unlocked AD107 GPU according to leak first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4060-will-use-an-unlocked-ad107-gpu-according-to-leak/feed/ 0
Latest Nvidia RTX 4070 rumours claim 2475MHz boost clock speed https://www.kitguru.net/components/graphic-cards/joao-silva/latest-nvidia-rtx-4070-rumours-claim-2475mhz-boost-clock-speed/ https://www.kitguru.net/components/graphic-cards/joao-silva/latest-nvidia-rtx-4070-rumours-claim-2475mhz-boost-clock-speed/#respond Wed, 01 Feb 2023 14:00:58 +0000 https://www.kitguru.net/?p=592021 Nvidia unveiled the RTX 4070 Ti last month at CES and it seems the RTX 4070 won't be too far behind. According to recent reports, board partners are currently in the process of preparing custom RTX 4070 graphics cards, and we may already know the GPU's base and boost clock speeds.  As per T4C Fantasy …

The post Latest Nvidia RTX 4070 rumours claim 2475MHz boost clock speed first appeared on KitGuru.]]>
Nvidia unveiled the RTX 4070 Ti last month at CES and it seems the RTX 4070 won't be too far behind. According to recent reports, board partners are currently in the process of preparing custom RTX 4070 graphics cards, and we may already know the GPU's base and boost clock speeds. 

As per T4C Fantasy (via VideoCardz), the Nvidia RTX 4070 should have a base clock speed of 1,920MHz and a boost clock speed of 2,475MHz, lower than the RTX 4070 Ti. Moreover, it has been reported that there would be two reference models of the RTX 4070, but it's unclear what the differences are between the two and if both will be available at the same time.

The leaker also claimed that an AIB partner has already decided on the specifications of its custom RTX 4070 graphics cards. That would be Colorful with its iGame Ultra White OC line, which reportedly has a boost clock of 2,505MHz, 30MHz more than the reference value.

Based on previous rumours, the RTX 4070 should come with an AD104 GPU with 5,888 CUDA cores and 12GB of GDDR6X memory running at 21Gbps across a 192-bit memory bus for a maximum bandwidth of 504GB/s. All that would come in a 200W TDP. The RTX 4070 cards are expected to enter mass production this month, so we could still see them available in Q1 2023.

Discuss on our Facebook page, HERE.

KitGuru says: Based on the rumoured specifications of the RTX 4070, the upcoming GPU will be slower than the RTX 4070 Ti. However, depending on pricing, this could be this generation's most successful RTX card.

The post Latest Nvidia RTX 4070 rumours claim 2475MHz boost clock speed first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/latest-nvidia-rtx-4070-rumours-claim-2475mhz-boost-clock-speed/feed/ 0
Nvidia RTX 4090 laptop GPU scores higher than a desktop RTX 3090 in Geekbench https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4090-laptop-gpu-scores-higher-than-a-desktop-rtx-3090-in-geekbench/ https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4090-laptop-gpu-scores-higher-than-a-desktop-rtx-3090-in-geekbench/#respond Mon, 02 Jan 2023 16:58:22 +0000 https://www.kitguru.net/?p=588377 Nvidia RTX 40 series laptop GPUs are starting to show up in the Geekbench database. We've already seen an RTX 4060 laptop GPU entry. Today, we've come with another entry, but this time from an RTX 4090 laptop GPU, which outperformed the RTX 3090 desktop card in the same benchmark. In the Geekbench entry shared …

The post Nvidia RTX 4090 laptop GPU scores higher than a desktop RTX 3090 in Geekbench first appeared on KitGuru.]]>
Nvidia RTX 40 series laptop GPUs are starting to show up in the Geekbench database. We've already seen an RTX 4060 laptop GPU entry. Today, we've come with another entry, but this time from an RTX 4090 laptop GPU, which outperformed the RTX 3090 desktop card in the same benchmark.

In the Geekbench entry shared by Benchleaks, we see an undisclosed laptop (X370SNX) equipped with an RTX 4090 laptop GPU, a 24-core Intel Core i9-13900HX and 32GB of memory. Moreover, the entry also reveals some of the specifications of the GPU, including that it has 76 CUs, meaning it has 9,728 CUDA cores. Lastly, it confirms the GPU will pack 16GB of VRAM and feature a maximum boost frequency of 2,040MHz.

The entry shows that the GPU scored 210,290 points in the OpenCL benchmark, ranking it 2.5% above your average RTX 3090. Comparing this score to the fastest RTX 30 series laptop GPU, the RTX 3080 Ti, the upcoming mobile GPU is 55% faster, a considerable generational performance uplift between flagship SKUs.

Currently, it is rumoured that Nvidia will unveil the first RTX 40-powered laptops at CES.

Kitguru says: Are you going to be on the lookout for a new gaming laptop next year? 

The post Nvidia RTX 4090 laptop GPU scores higher than a desktop RTX 3090 in Geekbench first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4090-laptop-gpu-scores-higher-than-a-desktop-rtx-3090-in-geekbench/feed/ 0
Nvidia RTX 4060 Ti to feature an AD106 GPU with 4,352 CUDA cores and 8GB of memory https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4060-ti-to-feature-an-ad106-gpu-with-4352-cuda-cores-and-8gb-of-memory/ https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4060-ti-to-feature-an-ad106-gpu-with-4352-cuda-cores-and-8gb-of-memory/#respond Thu, 15 Dec 2022 12:30:34 +0000 https://www.kitguru.net/?p=586601 Sooner or later, Nvidia will expand the RTX 40 series with mid-range SKUs. One of these SKUs should be the RTX 4060 Ti, which may have had its specs leaked this week.  Twitter leaker, kopite7kimi, has provided an update on AD106, the GPU expected to power the RTX 4060 Ti. According to the leaker, the …

The post Nvidia RTX 4060 Ti to feature an AD106 GPU with 4,352 CUDA cores and 8GB of memory first appeared on KitGuru.]]>
Sooner or later, Nvidia will expand the RTX 40 series with mid-range SKUs. One of these SKUs should be the RTX 4060 Ti, which may have had its specs leaked this week. 

Twitter leaker, kopite7kimi, has provided an update on AD106, the GPU expected to power the RTX 4060 Ti. According to the leaker, the upcoming mid-range graphics card has a “very short reference board” that uses the CEM5 connector (16-pin power).

As for the specs themselves, the RTX 4060 Ti is rumoured to feature an AD106-350 GPU with 4,352 CUDA cores and 32MB of L2 cache. Regarding the memory subsystem, it's expected to come with 8GB of GDDR6 memory clocked at 18Gbps. Past rumours claim the GPU has a 128-bit memory bus, but Nvidia might change to 256-bit one. All this comes in a 220W TDP, 20W more than the RTX 3060 Ti.

The Nvidia RTX 4060 Ti is expected to release sometime in 2023, probably after the RTX 4070 Ti. With some luck, Nvidia might give us a few more details about this GPU in its CES 2023 keynote.

Discuss on our Facebook page, HERE.

KitGuru says: Based on the leaked specifications, how do you think the RTX 4060 Ti will compare to the RTX 3060 TI?

The post Nvidia RTX 4060 Ti to feature an AD106 GPU with 4,352 CUDA cores and 8GB of memory first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4060-ti-to-feature-an-ad106-gpu-with-4352-cuda-cores-and-8gb-of-memory/feed/ 0
Intel details entry-level Arc A310 desktop GPU specs https://www.kitguru.net/components/graphic-cards/joao-silva/intel-shares-entry-level-arc-a310-desktop-gpu-specs/ https://www.kitguru.net/components/graphic-cards/joao-silva/intel-shares-entry-level-arc-a310-desktop-gpu-specs/#respond Fri, 30 Sep 2022 09:30:03 +0000 https://www.kitguru.net/?p=576195 Besides announcing the Arc A770 and A750 GPUs, both launching on October 12th, Intel has also given us a brief look at its upcoming entry-level GPU, the Arc A310. Based on its specifications, don't expect it to do battle with any of Nvidia or AMD's mid to high-end GPUs. However, there are use cases where …

The post Intel details entry-level Arc A310 desktop GPU specs first appeared on KitGuru.]]>
Besides announcing the Arc A770 and A750 GPUs, both launching on October 12th, Intel has also given us a brief look at its upcoming entry-level GPU, the Arc A310. Based on its specifications, don't expect it to do battle with any of Nvidia or AMD's mid to high-end GPUs. However, there are use cases where such a card can come in handy. 

The Intel Arc A310 will use a cut-down ACM-G11 GPU based on the TSMC N6 node, featuring just six Xe-cores clocked at 2000MHz. The memory subsystem has 4GB of GDDR6 memory running at 15.5Gbps across a 64-bit memory interface, offering a maximum bandwidth of 124GB/s. All this comes in a 75W TBP package, meaning we should see models without any power connectors.

According to the spec sheet, the new graphics card supports up to PCIe 4.0 x16. However, considering the A380 only supports eight lanes, we wouldn't be surprised if that was a typo.

Intel's new entry-level graphics card will be available in Q3 2022. Considering the quarter will end in just two days, Intel will likely miss the release date, as it did with the Arc A770.

Discuss on our Facebook page, HERE.

KitGuru says: For basic graphics tasks, the Arc A310 should do the trick. For anything beyond that, you'll likely be better off with a GPU that offers a little more power. 

The post Intel details entry-level Arc A310 desktop GPU specs first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/intel-shares-entry-level-arc-a310-desktop-gpu-specs/feed/ 0
Intel finally reveals full specs for its Arc Alchemist GPU lineup https://www.kitguru.net/components/graphic-cards/joao-silva/intel-finally-reveals-full-specs-for-its-arc-alchemist-gpu-lineup/ https://www.kitguru.net/components/graphic-cards/joao-silva/intel-finally-reveals-full-specs-for-its-arc-alchemist-gpu-lineup/#respond Fri, 09 Sep 2022 08:00:32 +0000 https://www.kitguru.net/?p=573850 After many months of speculation, rumours and reports, Intel has finally revealed what's under the hood of its yet unreleased Arc Alchemist Arc GPUs. We've known about the specifications of the A380 since it was first launched in China, but up until now we've been kept in the dark about the A580, A750 and A770...

The post Intel finally reveals full specs for its Arc Alchemist GPU lineup first appeared on KitGuru.]]>
After many months of speculation, rumours and reports, Intel has finally revealed what's under the hood of its yet unreleased Arc Alchemist Arc GPUs. We've known about the specifications of the A380 since it was first launched in China, but up until now we've been kept in the dark about the A580, A750 and A770.

According to Intel's information, the Arc A770 will feature a full ACM-G10 GPU with 32 Xe-Cores, 32 ray tracing units, 512 XMX engines, and a base clock of 2100MHz. The Arc flagship card will be available with 8/16GB of GDDR6 memory clocked at 17.5Gbps across a 256-bit memory bus, resulting in a maximum memory bandwidth of 560GB/s. However, the Limited Edition model will only be available with 16GB. All this comes in a 225W TDP design.

The A750 will feature a cut-down ACM-G10 GPU with 28 Xe-Cores, 28 ray tracing units, and 448 XMX engines. The base clock speed has also been reduced to 2050MHz and the memory speed to 16Gbps. With that speed and its 256-bit memory bus, the card's maximum memory bandwidth is 512GB/s. Unlike the A770, the A750 will only be available with 8GB of GDDR6 memory.

Lastly, the A580 will arrive with its GPU packing 24 Xe-Cores, 24 ray tracing units and 384 XMX engines. The base clock speed is set at a mere 1700MHz, while the 8GB of GDDR6 memory will run at 16Gbps across a 256-bit memory bus. That gives it the same maximum memory bandwidth as the A750, at 512GB/s.

Intel representatives recently told the media that the ARC A5 and A7 cards will release ‘very soon', but nothing has been officially confirmed just yet.

Discuss on our Facebook page HERE.

KitGuru says: Note that the clocks speeds shared by Intel are “base”, meaning the GPUs will boost much higher if it has the room to do so.

The post Intel finally reveals full specs for its Arc Alchemist GPU lineup first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/intel-finally-reveals-full-specs-for-its-arc-alchemist-gpu-lineup/feed/ 0
Intel Xeon “Emerald Rapids” will natively support DDR5-5600 memory according to leak https://www.kitguru.net/components/cpu/joao-silva/intel-xeon-emerald-rapids-will-natively-support-ddr5-5600-memory-according-to-leak/ https://www.kitguru.net/components/cpu/joao-silva/intel-xeon-emerald-rapids-will-natively-support-ddr5-5600-memory-according-to-leak/#respond Sun, 04 Sep 2022 11:20:45 +0000 https://www.kitguru.net/?p=573027 While we wait for the Intel Xeon “Sapphire Rapids” platform to see the light of day, reports about its successor have started to emerge. Named “Emerald Rapids”, the upcoming platform will be much like its predecessor, with slight improvements in memory support, maximum core count, and security. Leaker @yukki_ans has recently shared multiple screenshots showing …

The post Intel Xeon “Emerald Rapids” will natively support DDR5-5600 memory according to leak first appeared on KitGuru.]]>
While we wait for the Intel Xeon “Sapphire Rapids” platform to see the light of day, reports about its successor have started to emerge. Named “Emerald Rapids”, the upcoming platform will be much like its predecessor, with slight improvements in memory support, maximum core count, and security.

Leaker @yukki_ans has recently shared multiple screenshots showing details about the upcoming Xeon series. Like Sapphire Rapids, Emerald Rapids will be based on Intel 7 architecture. It will also make use of the same Eagle Stream platform, with TDPs ranging between 125W and 350W, 80 PCIe lanes, eight DDR5 memory channels, and scalability for up to eight sockets,

However, Emerald Rapids will also bring improvements. The Sapphire Rapids platform is limited to DDR5-4800 in 1DPC (DIMM per channel) and 2DPC configurations, but Emerald Rapids isn't. In 2DPC configurations, the latter supports the same memory speeds as the former, but in 1DPC, Emerald Rapids will natively support DDR5-5600.

UPI 2.0 (up to four links) support is also improving. While Sapphire Rapids will be limited to 16GT/s speeds, Emerald Rapids will be able to go up to 20GT/s. Lastly, Intel 5th Gen Xeon processors will feature Trust Domain Extensions, which lets users deploy hardware-isolated VMs. Despite the various delays to Sapphire Rapids, Emerald Rapids chips are still set to release in late 2023.

KitGuru says: These Xeons will likely be going up against a new generation of Zen 4 EPYC processors, so it’ll be interesting to see comparisons when the time comes.  

The post Intel Xeon “Emerald Rapids” will natively support DDR5-5600 memory according to leak first appeared on KitGuru.]]>
https://www.kitguru.net/components/cpu/joao-silva/intel-xeon-emerald-rapids-will-natively-support-ddr5-5600-memory-according-to-leak/feed/ 0
Nvidia RTX 4080 to feature less CUDA cores than initially expected https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4080-to-feature-less-cuda-cores-than-initially-expected/ https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4080-to-feature-less-cuda-cores-than-initially-expected/#respond Mon, 08 Aug 2022 16:00:20 +0000 https://www.kitguru.net/?p=570204 There have been a number of rumours regarding Nvidia's next generation RTX graphics cards. In the latest round-up, one source changed their expected spec for the RTX 4070 and now, specifications for the upcoming RTX 4080 have also been adjusted, featuring less CUDA cores than previously expected.  According to kopite7kimi, an account that has been …

The post Nvidia RTX 4080 to feature less CUDA cores than initially expected first appeared on KitGuru.]]>
There have been a number of rumours regarding Nvidia's next generation RTX graphics cards. In the latest round-up, one source changed their expected spec for the RTX 4070 and now, specifications for the upcoming RTX 4080 have also been adjusted, featuring less CUDA cores than previously expected. 

According to kopite7kimi, an account that has been sharing updates on Nvidia's new-gen GPUs and has made a number of adjustments along the way, the RTX 4080 will now feature fewer CUDA cores than previously stated. According to their information, the RTX 4080 will feature 9,728 CUDA Cores, down from the initial 10,240 CUDA Core estimate. It is also claimed that these graphics cards will have a 420W TDP, an AD103-300-A1 GPU, 16GB of GDDR6X memory and a PG136/139-SKU360 board.

While there has been a 5% reduction in CUDA cores, performance shouldn't be impacted too much. It is still claimed that the RTX 4080 should be capable of scoring around 15K points in the TimeSpy benchmark. Depending on power draw, it might also be feasible to gain that performance back through overclocking.

The RTX 40 series is expected to be announced in Q4, starting with the RTX 4090, RTX 4080 and RTX 4070, although the release of these graphics cards will likely be staggered.

KitGuru says: These are all rumoured specifications for now and we can't verify the claims made. With that in mind, treat all information as rumour while we await a proper announcement from Nvidia. 

The post Nvidia RTX 4080 to feature less CUDA cores than initially expected first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/nvidia-rtx-4080-to-feature-less-cuda-cores-than-initially-expected/feed/ 0
AMD Ryzen 7000 processors to feature up to 16 cores and boost up to 5.7GHz https://www.kitguru.net/components/cpu/joao-silva/amd-ryzen-7000-processors-to-feature-up-to-16-cores-and-boost-up-to-5-7ghz/ https://www.kitguru.net/components/cpu/joao-silva/amd-ryzen-7000-processors-to-feature-up-to-16-cores-and-boost-up-to-5-7ghz/#respond Fri, 05 Aug 2022 15:50:34 +0000 https://www.kitguru.net/?p=570136 There's still much to learn about the Ryzen 7000 series. The official announcement is expected to happen later this month, but while we wait for it, leakers have shared what they claim to be the specifications of AMD's next-gen Ryzen CPUs. According to the information obtained by Wccftech, the Ryzen 7000 series will launch with …

The post AMD Ryzen 7000 processors to feature up to 16 cores and boost up to 5.7GHz first appeared on KitGuru.]]>
There's still much to learn about the Ryzen 7000 series. The official announcement is expected to happen later this month, but while we wait for it, leakers have shared what they claim to be the specifications of AMD's next-gen Ryzen CPUs.

According to the information obtained by Wccftech, the Ryzen 7000 series will launch with four processors – the Ryzen 9 7950X, Ryzen 9 7900X, Ryzen 7 7700X and Ryzen 5 7600X. These CPUs will come with up to 16 Zen 4 cores capable of reaching 5.7GHz boost clock speeds and packing up to 80MB of cache (L3+ L2). As per the rumours, the maximum TDP of the upcoming chips is 170W.

The following table shows the rumoured specifications for these SKUs in more detail:

CPU Cores/ Threads Base Clock Boost Clock Cache (L3+ L2) TDP
AMD Ryzen 9 7950X 16/32 4.5 GHz 5.7 GHz 80 MB (64+16) 170W
AMD Ryzen 9 7900X 12/24 4.7 GHz 5.6 GHz 76 MB (64+12) 170W
AMD Ryzen 7 7700X 8/16 4.5 GHz 5.4 GHz 40 MB (32+8) 105W
AMD Ryzen 5 7600X 6/12 4.7 GHz 5.3 GHz 38 MB (32+6) 105W

AMD has already introduced the first X670 and X670E motherboards, so all that's left is for AMD to announce the Ryzen 7000 series. Recent rumours claim we'll see these CPUs announced on the 29th of August, with retail availability arriving in mid-September.

Discuss on our Facebook page, HERE.

KitGuru says: Based on these specifications, which Ryzen 7000 processor looks the best for your needs? Do you think Intel Raptor Lake will offer a compelling alternative to AMD's next-gen CPUs?

The post AMD Ryzen 7000 processors to feature up to 16 cores and boost up to 5.7GHz first appeared on KitGuru.]]>
https://www.kitguru.net/components/cpu/joao-silva/amd-ryzen-7000-processors-to-feature-up-to-16-cores-and-boost-up-to-5-7ghz/feed/ 0
Rumour claims RTX 4070 to be as powerful as RTX 3090 Ti https://www.kitguru.net/components/graphic-cards/joao-silva/rumour-claims-rtx-4070-to-be-as-powerful-as-rtx-3090-ti/ https://www.kitguru.net/components/graphic-cards/joao-silva/rumour-claims-rtx-4070-to-be-as-powerful-as-rtx-3090-ti/#respond Fri, 05 Aug 2022 15:15:34 +0000 https://www.kitguru.net/?p=570135 Although they're not official, many were let down by the rumoured specifications of the RTX 4070, with early rumours claiming the card would feature 10GB of GDDR6 memory across a 160-bit memory bus. However, a more recent report suggests the upcoming SKU might end up faster than initially expected. The latest report regarding the RTX …

The post Rumour claims RTX 4070 to be as powerful as RTX 3090 Ti first appeared on KitGuru.]]>
Although they're not official, many were let down by the rumoured specifications of the RTX 4070, with early rumours claiming the card would feature 10GB of GDDR6 memory across a 160-bit memory bus. However, a more recent report suggests the upcoming SKU might end up faster than initially expected.

The latest report regarding the RTX 4070 posted by kopite7kimi claims the graphics card may use the PG141-SKU331 with a fully enabled AD104 GPU. This SKU was revealed a few days ago, but back then, it was unclear how it would fit within the RTX 40 series. Now that more information has come to light, it looks like it will be a replacement for the PG141-SKU341, thought to be used on the RTX 4070.

With a full AD104 GPU, the number of CUDA cores of the RTX 4070 increased from 7,168 to 7,680. Moreover, the memory configuration was upgraded too, now listed with 12GB of GDDR6X at 21Gbps on a 192-bit memory bus. Despite all the upgrades, the TDP is expected to remain the same, rumoured to be set at 300W.

Nvidia has not officially announced a new generation of RTX graphics cards yet or confirmed any details. However, it is expected that new Nvidia graphics cards will launch later this year.

Discuss on our Facebook page, HERE.

KitGuru says: For now, this is all rumour, but there is no denying that getting x90Ti level performance in an x70 series card would be an impressive generational jump. 

The post Rumour claims RTX 4070 to be as powerful as RTX 3090 Ti first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/rumour-claims-rtx-4070-to-be-as-powerful-as-rtx-3090-ti/feed/ 0
Nvidia’s flagship next-gen GPU may come with 48GB of GDDR6X memory https://www.kitguru.net/components/graphic-cards/joao-silva/nvidias-flagship-next-gen-gpu-may-come-with-48gb-of-gddr6x-memory/ https://www.kitguru.net/components/graphic-cards/joao-silva/nvidias-flagship-next-gen-gpu-may-come-with-48gb-of-gddr6x-memory/#respond Tue, 26 Jul 2022 16:00:31 +0000 https://www.kitguru.net/?p=569099 There are plenty of rumours swirling around Nvidia's next-generation of GeForce RTX graphics cards. We've already heard several claims about the RTX 4090, RTX 4080 and RTX 4070. According to a new rumour, there will be an even more powerful graphics card based on the AD102 chip.  According to kopite7kimi, Nvidia is working on a …

The post Nvidia’s flagship next-gen GPU may come with 48GB of GDDR6X memory first appeared on KitGuru.]]>
There are plenty of rumours swirling around Nvidia's next-generation of GeForce RTX graphics cards. We've already heard several claims about the RTX 4090, RTX 4080 and RTX 4070. According to a new rumour, there will be an even more powerful graphics card based on the AD102 chip. 

According to kopite7kimi, Nvidia is working on a graphics card based on the AD102-450-A1 GPU, featuring 18,176 CUDA cores. This card will apparently use a different board to the RTX 4090, as well as an upgraded memory subsystem with 48GB of GDDR6X memory clocked at 24Gbps. It is also claimed that the graphics card may have an 800W TBP.

Judging by the specs, if this GPU does exist, it may well be a next generation Titan, or perhaps an RTX 4090 Ti. We are likely to see other RTX 40 series graphics cards first, with rumours claiming a 2022 launch for the RTX 4090, followed by the RTX 4080 and RTX 4070.

Discuss on our Facebook page, HERE.

KitGuru says: With an 800W TBP, most people would likely need a PSU upgrade before running one of these in their system. Nothing has been officially announced yet though, so as always, treat this information as rumour rather than fact. 

The post Nvidia’s flagship next-gen GPU may come with 48GB of GDDR6X memory first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/nvidias-flagship-next-gen-gpu-may-come-with-48gb-of-gddr6x-memory/feed/ 0
OEM RTX 3050 may have fewer CUDA cores than retail version https://www.kitguru.net/components/graphic-cards/joao-silva/oem-rtx-3050-may-have-fewer-cuda-cores-than-retail-version/ https://www.kitguru.net/components/graphic-cards/joao-silva/oem-rtx-3050-may-have-fewer-cuda-cores-than-retail-version/#respond Tue, 19 Jul 2022 15:30:35 +0000 https://www.kitguru.net/?p=568397 It seems Nvidia is giving a different RTX 3050 GPU with fewer CUDA cores to OEMs. The specifications of the new RTX 3050 variant were made public through Nvidia's official website, allowing us to see the differences between both the standard and OEM variants. Weibo user Golden Pig Upgrade Pack shared a photo of what …

The post OEM RTX 3050 may have fewer CUDA cores than retail version first appeared on KitGuru.]]>
It seems Nvidia is giving a different RTX 3050 GPU with fewer CUDA cores to OEMs. The specifications of the new RTX 3050 variant were made public through Nvidia's official website, allowing us to see the differences between both the standard and OEM variants.

Weibo user Golden Pig Upgrade Pack shared a photo of what seems to be the OEM version of the RTX 3050 graphic card. According to the uploader, this version of the GPU only features 2,304 CUDA cores, a considerable cut from the 2,560 CUDA cores of the reference model. As per the photo, the RTX 3050 for OEMs is powered by an 8-pin power connector and features a dual-fan cooling design. The PCB looks similar to the one used in the reference model.


Image credit: Weibo

As it seems, this card is relatively new. Nvidia has recently updated the spec summary of the RTX 3050, adding the 2,304 CUDA cores variant. According to the spec sheet, this GPU also comes with slightly lower clock speeds, as the GPU has a 1.51GHz base clock and a 1.76 GHz boost clock. TDP (130W) and memory configuration (8GB GDDR6 at 14Gbps on a 128-bit memory bus) remain unchanged.

We don't know what GPU revision is being used on the RTX 3050 OEM, but VideoCardz estimates that it may be the GA107-140 revision, which we heard about earlier this year.

Discuss on our Facebook page, HERE.

KitGuru says: Considering it features fewer CUDA cores and lower clock speeds, this version of the RTX 3050 should also draw less power than the standard model. Unfortunately, Nvidia's spec sheet shows that's not the case.

The post OEM RTX 3050 may have fewer CUDA cores than retail version first appeared on KitGuru.]]>
https://www.kitguru.net/components/graphic-cards/joao-silva/oem-rtx-3050-may-have-fewer-cuda-cores-than-retail-version/feed/ 0