Intel's processor plans for 2023 have been revealed: Raptor Lake-S Refresh, Sapphire Rapids-WS and special Sapphire Rapids-SP for workstations

Intel is going to hold a webinar for investors next week, where it will reveal the current plans for various products.
This information has been leaked to the Net ahead of time.
The manufacturer is going to present next year updated consumer Raptor Lake, chips for workstations Sapphire Rapids-WS, as well as special Sapphire Rapids-SP for workstations.
All of the new products will be released over the next five quarters.Image source: IntelJudging from the published presentation slides, the updated Raptor Lake-S Refresh series of consumer processors will be released early in the third quarter of next year, which is less than a year after the first Intel Core 13th generation processor models were released.
Intel Raptor Lake-S Refresh.
Image source: Twitter / @9550proAccording to Intel's consumer processor roadmap, the company will refresh all of its 13th-generation Core models with TDP of 125, 65 and 35 watts.
Notably, the manufacturer has not yet released Raptor Lake processor models with a nominal TDP below 125W at this time.
Their announcement is expected only in early January next year.
Intel chipset roadmapIt is also important to note that for the updated Raptor Lake-S Refresh processors, the manufacturer does not plan to release a new series of chipsets for motherboards.
In other words, all future upgraded Raptor Lake models will be able to run on motherboards with Intel 700-series chipsets.
To date, only the elder Intel Z790 series chipset is available in this series.
Also manufacturer will present chipsets H770 and B760, but in the entry-level segment will remain the already released H610.
Features of Sapphire Rapids-WS (Intel Xeon W-3400)Within Sapphire Rapids-WS platform for workstations Intel will release two series of processors: Xeon W-3400 and Xeon W-2400.
The former will be able to offer from 12 to 56 cores with support for 24 to 112 virtual threads, and will run at up to 4.8 GHz.
These processors will get 30 to 105 Mbytes of L3 cache and will have a TDP rating of 220 to 350 watts.
Intel Xeon W-3400 lineupSome of the Intel Xeon W-3400 processors will have an unlocked multiplier, allowing for manual overclocking.
It will support up to 4TB of 8-channel DDR5-4400 and DDR5-4800 RAM, as well as 112 lanes of PCIe 5.0.
Features of Sapphire Rapids-WS (Intel Xeon W-2400)Xeon W-2400 series will be represented by chips with the number of cores from 6 to 24, which will support from 12 to 48 virtual threads.
They will have 15 to 45 Mbytes of L3 cache memory and will run at speeds up to 4.8 GHz.
The TDP rating of these processor models will range from 110W to 225W.
Intel Xeon W-3400 lineup Part of the Xeon W-2400 series will also get an unlocked multiplier.
These processors will support up to 2TB of quad-channel DDR5-4400 and DDR5-4800 RAM, as well as 64 lanes of PCIe 5.0.
Features of the Sapphire Rapids-SP platform for workstationsSapphire Rapids-SP server-class processors will be represented by a series of 4th generation Xeon Scalable Platinum, Gold, Silver and Bronze models.
The Xeon Scalable Platinum Series chips will offer 36 to 56 processing cores with support for 72 to 112 virtual threads.
These chips will not have support for Intel Boost Max 3.0 technology, which will be available on Sapphire Rapids-WS workstation models, and therefore will have a maximum clock speed of 3.8 GHz.
This platform will get support for 80 PCIe 5.0 lanes and up to 4TB of DDR5-4800 RAM per CPU slot.
They will have a TDP rating of 300 to 350W.
Intel Sapphire Rapids-SP Workstation (Xeon Platinum 8400)Gold models will offer 8 to 32 cores with support for 16 to 64 virtual threads and will run at clock speeds up to 4.1 GHz.
Those processors will range from 22.5 to 60 Mbytes of L3 cache and have a TDP from 185W to 250W, depending on model.
The amount and standard of supported DDR5 memory, as well as the number of available PCIe 5.0 lanes will be the same as in Platinum models.
Intel Sapphire Rapids-SP for workstations (Xeon Gold 6400)Silver series will feature two CPU models: with 12 and 20 cores, supporting 24 or 40 threads respectively.
Both chips will run at 3.9 GHz.
The processors will get 30 and 37.5 Mbytes of L3 cache, respectively.
Their TDP will be 150 and 165W respectively.
Intel Sapphire Rapids-SP workstation lineup (Xeon Silver 4400 and Bronze 3400)Bronze lineup will feature only one 8-core model, without multithreading support

The circuit board of the reference GeForce RTX 4080 16GB Founders Edition showed up in the photo

Photos of the circuit board of the reference graphics card GeForce RTX 4080 16GB Founders Edition have been published on the web.
They show a large-scale AD103 GPU on the Ada Lovelace architecture.
Twitter user @KittyYYuko shared the pictures.
Image source: Twitter / @KittyYYuko In an effort to hide details that could tip off NVIDIA as the informant, @KittyYYuko covered some parts of the PCB with an Intel processor and an EK sticker.
Unfortunately, the presence of these items on the PCB makes it impossible to estimate, the exact number of power phases used by the reference version of the GeForce RTX 4080.
However, we can note that some of the slots for the elements of the VRM subsystem are empty at all.
This may hint that NVIDIA in the future may release a more powerful solution, for example, in the form of GeForce RTX 4080 Ti, which will then undoubtedly have a more efficient power subsystem.
Officially, NVIDIA has not said anything about such plans yet.
On the photo of the PCB you can also note the presence of the latter 12+4-pin power connector 12VHPWR.
Apparently, this standard will be used in all senior models of GeForce RTX 40-series gas pedals.
Recall that at the core of the GeForce RTX 4080 16GB uses a graphics processor with 9728 CUDA cores.
The card has 16 GB of GDDR6X memory.
The photo shows that the gas pedal is equipped with eight Micron memory modules of 2 GB each.
The memory is connected to the GPU via a 256-bit bus.The informant also said that the card scored about 14,100 graphics performance points in the 3DMark Time Spy Extreme synthetic test, which is also confirmed by earlier leaks from other insiders.
Screenshots of the latest such leak can be seen below.
Source image: Twitter / MEGAsizeGPUThe GeForce RTX 4080 16GB sales launch is scheduled for November 16.
The recommended price of the gas pedal in the Founders Edition reference version is $1199.

AMD Navi 31 flagship multi-chip GPU showed up in photos ahead of the announcement

A photo of AMD Navi 31 GPU based on RDNA 3 architecture, which will form the basis of higher-end Radeon RX 7000 graphics cards appeared at Angstronomics' disposal.
Recall that the official announcement of the latter will be held this evening.Image source: AMDThe presented photo of Navi 31 GPU confirms the use of chiplet design.
In the photo you can note the main graphics chip Graphics Chiplet Die (GCD) in the center and six auxiliary crystals Memory Chiplet Dies (MCD).Image source: AngstronomicsThe same source previously reported the alleged specifications of the Navi 31 GPU.
So, the main GCD chip with an area of about 308 mm2 contains 48 WGP (Workgroup Processors) blocks.
Each WGP has two Compute Units (CUs).
Thus, the total number of CUs in the chip is 96.
And the total number of Arithmetic Logic Units (ALUs or Area Stream Processors) within the 96 CU's would be equal to 12,288 units.The area of one MCD chip, in turn, is about 37.5 mm2.
The total area of the GCD with all the MCD chips is approximately 533 mm2.
The area of the entire GPU package is 2612 mm2 (dimensions 55 × 47.5 mm).
Each MCD contains 16 Mbytes of AMD Infinity Cache.
Thus, Navi 31 GPU will get 96 MB of Infinity Cache, which is less than the same Navi 21 (128 MB) that powers Radeon RX 6900 XT/6950 XT.It is also reported that Navi 31 GPU supports 384-bit memory bus with up to 24 GB of GDDR6 memory at 20 Gbit/s per pin.
This chip is expected to be the core of the Radeon RX 7900 XTX graphics gas pedal.
Expected specifications of AMD Navi 31, Navi 32 and Navi 33 graphics processors.
Image source: VideoCardzIt is expected that during today's presentation AMD will present only the senior models of Radeon RX 7000 gas pedals based on Navi 31.
Models based on Navi 32 and Navi 33 GPUs will probably be announced later.
However, we will be able to find out the exact information in a few hours.

Radeon RX 7900 in reference version showed up in photos - card has two 8-pin power connectors

One of Twitter users shared photos of reference version of AMD Radeon RX 7900 graphics gas pedal.
According to him, the pictures were published in one of the closed communities for enthusiasts.\t\t\t\t\t\t\t\tImage source: Twitter / @9550pro The photo does not show the final version of the graphics gas pedal, but only an engineering sample.
Such a conclusion can be made at least from the fact that the card uses a red PCB, on which in some places you can see contact points for voltage monitoring.
However, the gas pedal cooling system is very similar to the one that the final version will receive.
Its outlines could be seen back at the August AMD presentation.In the photos presented Radeon RX 7900 is compared with the reference version Radeon RX 6900 XT.
Novelty is longer than its predecessor but has the same set of two 8-pin power connectors as the model on the RDNA 2 architecture.
The appearance of cooling system has changed relative to the previous generation graphics cards Radeon.
However, these changes are more of a cosmetic nature.\t\t\t\tThe cooling system of the reference version of the new generation AMD video card includes three fans.
The source did not specify what gas pedal model on the RDNA 3 architecture has got on the photo.
It could be either Radeon 7900 XT, or the flagship Radeon RX 7900 XTX.
Below in the table you can see the characteristics that are attributed to these gas pedals.
In particular, GPUs of video cards will allegedly get up to 12,288 stream processors, and video cards themselves will be equipped with up to 24 GB of GDDR6 memory.\t\t\t\t\t\t\t\tImage source: VideoCardz Video cards Radeon RX 7000 series will be officially presented on November 3.
According to recent rumors, they will go on sale in early December.
We expect the announcement of at least two models of gas pedals based on RDNA 3 architecture - the aforementioned Radeon 7900 XT and Radeon RX 7900 XTX based on Navi 31 processors.
If you believe the rumors, the older models Radeon RX 7000 will use the GPU with an innovative chip design.

Radeon RX 7900 in reference version showed up in photos - card has two 8-pin power connectors

One of Twitter users shared photos of reference version of AMD Radeon RX 7900 graphics gas pedal.
According to him, the pictures were published in one of the closed communities for enthusiasts.\t\t\t\t\t\t\t\tImage source: Twitter / @9550pro The photo does not show the final version of the graphics gas pedal, but only an engineering sample.
Such a conclusion can be made at least from the fact that the card uses a red PCB, on which in some places you can see contact points for voltage monitoring.
However, the gas pedal cooling system is very similar to the one that the final version will receive.
Its outlines can be seen back at the August presentation of AMD.In the photos presented Radeon RX 7900 is compared with the reference version of the Radeon RX 6900 XT.
Novelty is longer than its predecessor but has the same set of two 8-pin power connectors as the model on the RDNA 2 architecture.
The appearance of cooling system has changed relative to the previous generation graphics cards Radeon.
However, these changes are more of a cosmetic nature.\t\t\t\tThe cooling system of the reference version of the new generation AMD video card includes three fans.
The source did not specify what gas pedal model on the RDNA 3 architecture has got on the photo.
It could be either Radeon 7900 XT, or the flagship Radeon RX 7900 XTX.
Below in the table you can see the characteristics that are attributed to these gas pedals.
In particular, GPUs of video cards will allegedly get up to 12,288 stream processors, and video cards themselves will be equipped with up to 24 GB of GDDR6 memory.\t\t\t\t\t\t\t\tImage source: VideoCardz Video cards Radeon RX 7000 series will be officially presented on November 3.
According to recent rumors, they will go on sale in early December.
We expect the announcement of at least two models of gas pedals based on RDNA 3 architecture - the aforementioned Radeon 7900 XT and Radeon RX 7900 XTX based on Navi 31 processors.
If you believe the rumors, the older models Radeon RX 7000 will use the GPU with an innovative chip design.

MSI's GeForce RTX 4080 in Suprim X, Gaming X Trio and Ventus 3X versions showed in images

TechPowerUp has got images of GeForce RTX 4080 graphics cards from MSI.
The manufacturer is going to release at least three versions, each in two variants with different frequencies: GeForce RTX 4080 Suprim X and Suprim, GeForce RTX 4080 Gaming X and Gaming, as well as GeForce RTX 4080 Ventus 3X OC and Ventus 3X.
A total of six models.Image source: MSI / TechPowerUpThe GeForce RTX 4080 Ventus 3X gas pedal will adhere to the stated reference GPU frequencies of NVIDIA itself, according to TechPowerUp.
The RTX 4080 Ventus 3X OC model will receive a slight factory overclocking of the GPU.
The flagship GeForce RTX 4080 Suprim X will offer the most significant factory GPU overclocking.
The GPU of the GeForce RTX 4080 Gaming X version will be overclocked to the level of the Suprim variant.
The GeForce RTX 4080 Gaming version, in turn, will run at frequencies higher than the Ventus 3X OC version.
At the same time, the card will offer a significantly more efficient cooling system.MSI GeForce RTX 4080 Suprim X/SuprimThe GeForce RTX 4080 Suprim X and Suprim cards look very similar to the GeForce RTX 4090 in the same design.
This means that their cooling systems are 4 expansion slots thick.
Despite the lack of flashy design, these new products will offer the most advanced technical equipment.
The mid-range GeForce RTX 4080 Gaming X and Gaming variants will also offer powerful cooling systems and plenty of RGB lighting.MSI GeForce RTX 4080 Gaming X/GamingThe GeForce RTX 4080 Ventus 3X OC and Ventus 3X are designed with practicality in mind.
In terms of technical characteristics and performance cards will be closest to the reference version of the GeForce RTX 4080.
And the main difference between them will be only in cooling systems.MSI GeForce RTX 4080 Ventus 3X OC/Ventus 3XWe should remind you that the GeForce RTX 4080 is scheduled to launch on November 16.
NVIDIA has priced the graphics card at $1199.

Ryzen 5 7725HS mobile processor on Zen 2 architecture has marked in Geekbench

The eight-core and 16-thread Ryzen 5 7725HS mobile processor on Zen 2 architecture has marked in Geekbench synthetic test database.
The software has determined that the chip belongs to a series of Renoir, writes VideoCardz.
The name of the processor also indicates that it is based on the architecture of Zen 2.
This is interesting, because AMD has not talked about the appearance of a similar model in the mobile Ryzen 7000 series.Image source: VideoCardz According to Geekbench data, the Ryzen 5 7725HS processor has a base frequency of 2.9 GHz and can automatically overclock to 4.3 GHz.
The suffix HS in the name of the chip indicates that it is designed for light gaming laptops.
The processor was tested with the Eluktronics RP-15 laptop without a discrete graphics subsystem.
In the single-threaded performance test Ryzen 5 7725HS scored 1106 points, in multi-threaded - 6929 points.
Image source: GeekBenchThe results of Ryzen 7 7725HSAs VideoCardz writes, it would be fair to compare this chip with the Ryzen 7 4800HS Renoir series, which also uses the Zen 2 architecture and eight processing cores.
In addition, the latter also has a TDP at 35W, like the novelty.
The obvious difference between the chips is the maximum clock speed, which is 4.2 GHz in the Ryzen 7 4800HS, which is 100 MHz lower.
Taking all this into account, the Ryzen 5 7725HS is 6 and 6.5% faster in single-threaded and multi-threaded performance, respectively, than the Ryzen 7 4800HS.

AMD will release 10-core Ryzen 7 7800X and 4-core Ryzen 3 7300X - new products have made a mark in Geekbench test

The Ryzen 7000 processor family should be updated with several new models in the foreseeable future.
Yesterday it was reported that data about Ryzen 7 7700 model appeared in Geekbench synthetic test database.
Now in the same database marked the Ryzen 7 7800X and Ryzen 3 7300X models, which are also not included in the first wave of AMD desktop processors on the architecture Zen 4.Image source: AMDThe Ryzen 7 7800X model has two more cores compared to its predecessor Ryzen 7 5800X.
Thus, the novelty could become the first AMD Ryzen desktop processor with 10 cores and 20 threads.
According to Geekbench, the chip can automatically overclock to 5.4 GHz and has 64 MB of L3 cache memory on board, which is twice as much as the Ryzen 7 7700X.
But that's not surprising, since the 10 cores are recruited using two chipsets, while the Ryzen 7 7700X has only one octa-core chip.
Image source: GeekbankshThe processor was tested on a Gigabyte Aorus B650 motherboard.
The Ryzen 7 7800X scored 2,097 points in the single-threaded performance test and 16,163 points in the multi-threaded test.
In terms of multi-threaded performance, the novelty in its current state is about 15% ahead of the Ryzen 7 7700X.Ryzen 7 7800X results.
Image source: GeekbenchThe Geekbench database also revealed information about the quad-core and eight-threaded Ryzen 3 7300X model.
According to the benchmark, this processor is able to automatically overclock up to 5.0 GHz and has 32 MB of L3 cache memory.
Image source: GeekbankshThe quad-core and octa-core Ryzen 3 7300X in the single-core test scored 1984 points, and in the multi-threaded performance test - 7682 points.
This chip is capable of delivering a clock speed of 5GHz in auto overclocking.Ryzen 3 7300X results.
Source image: GeekbenchWe should keep in mind that the Geekbench tests most likely involved engineering samples of processors, which may differ in frequency characteristics and power consumption from the final consumer versions.