#nvidia_graphics_cards

[phonemantra-blog]

Parameters are still unknown As you know, new US sanctions prohibit the supply of various accelerators for AI to China, including the H800 and A800 models, which, in fact, were created for China. However, Nvidia is simply going to pull the same trick again with the creation of special versions that will not fall under the next sanctions. [caption id="attachment_81201" align="aligncenter" width="600"] Nvidia[/caption] HGX H20, L20 PCIe and L2 PCIe models expected soon Nvidia is reportedly preparing three accelerators at once: HGX H20, L20 PCIe and L2 PCIe. They can be presented in the very near future: after November 16. It is still unknown what characteristics the new accelerators will have, but it is worth remembering that the new sanctions limit the supply of solutions with performance above a certain level, so Nvidia will clearly somehow limit this very performance. Let us also recall that the gaming RTX 4090 was also banned, but at least Nvidia has other options within the same line, and new ones will appear soon .

[phonemantra-blog]

The company is developing its own Ascend AI chips US measures to limit the export of advanced chips with artificial intelligence (AI) technologies to China could open up huge opportunities for Huawei for import substitution in the domestic market, as reported by Reuters. While Nvidia has historically been the leading AI chip supplier in China with a market share exceeding 90%, Chinese firms including Huawei are developing their versions of Nvidia's best-selling chips, such as the A100 and H100. Huawei's Ascend AI chips are comparable to Nvidia's in terms of processing power, according to analysts and some artificial intelligence companies such as China's iFlytek. [caption id="attachment_72981" align="aligncenter" width="600"] Nvidia[/caption] US ban on supplies of Nvidia accelerators opens up huge opportunities for Huawei Jiang Yifan, chief market analyst at brokerage Guotai Junan Securities, said another key limiting factor for Chinese firms is the dependence of most projects on Nvidia's chips and software ecosystem, but that could change with U.S. restrictions. “This US move, in my opinion, actually gives Huawei Ascend processors a huge gift,” Jiang said in a post on the social network Weibo. Many cutting-edge AI projects are being built using CUDA, a popular programming architecture pioneered by Nvidia, which in turn has led to the creation of a huge global ecosystem capable of training highly complex AI models such as OpenAI's GPT-4. Huawei's version is called CANN, and analysts say it is much more limited in terms of the artificial intelligence models it can train. Woz Ahmed, a former chip design executive turned consultant, said that for Huawei to win Chinese customers from Nvidia, it needs to replicate the ecosystem that Nvidia has created, including supporting customers in migrating their data and models to Huawei's platform. Woz Ahmed believes that Huawei could take 5 to 10 years to achieve this.

[phonemantra-blog]

There is a place in the lineup for them Yesterday the Internet started talking again about GeForce RTX 40 Super video cards, and now the characteristics of one of the models have appeared.  [caption id="attachment_71729" align="aligncenter" width="780"] Nvidia[/caption] The RTX 4070 Super, according to insider MEGAsizeGPU, will receive an AD103 GPU and 16 GB of memory with a 256-bit bus. This means that the new product will offer more memory than the RTX 4070 and RTX 4070 Ti, which, we recall, have 12 GB each. In addition, the use of the AD103 GPU hints that the RTX 4070 Super will be a step above the RTX 4070 Ti since the latter is based on the smaller AD104.   Nvidia is preparing two new video cards The insider also says that Nvidia may release a simplified version of the RTX 4070, which will use GDDR6 instead of GDDR6X memory, that is, the memory frequency will be lower. Whether there will be any other differences is still unclear, but such a model will close a rather large hole in terms of performance between the RTX 4070 and RTX 4060 Ti. True, let us remind you that the first costs $600, and the second in the 16 GB version costs $500, so it is not very clear how much a simplified RTX 4070 could cost so as not to destroy sales of the RTX 4060 Ti 16GB.  Insider MEGAsizeGPU previously spoke about the launch of the RTX 4060 and RTX 4060 Ti before the announcement, as well as the release dates for these models.

[phonemantra-blog]

So far only Radeon RX 460 You can connect a full desktop discrete graphics card   to a Raspberry Pi 5 single board computer for $60. [caption id="attachment_85125" align="aligncenter" width="780"] GeForce RTX 4090 to Raspberry Pi 5[/caption] This is exactly what channel author Jeff Geerling implemented by connecting a Radeon RX 460 adapter to his Pi 5. And it all worked. True, achieving this is not so easy. Of course, the Raspberry Pi 5 itself does not have a PCIe slot, and indeed there are no classic slots at all. But since this is a Raspberry Pi, its capabilities can be expanded with various modules, adapters, and other things. That is, you can even connect a GeForce RTX 4090 to Raspberry Pi 5? To connect the video card, the author of the project used the uPCity module, which is not yet commercially available, but will appear in time. In addition, of course, the card needed to implement additional power. https://youtu.be/BLg-1w2QayU It’s difficult to say whether someone needs such a solution for practical purposes, since the Raspberry Pi 5 was created for experiments, but the ability to connect a discrete video card in itself is already a big plus.

[phonemantra-blog]

But the configuration can be changed The GeForce RTX 4090 D (Dragon) video card , which will be created specifically for the Chinese market and will not be subject to US sanctions, will receive the same GPU as the current version.  [caption id="attachment_85108" align="aligncenter" width="600"] GeForce RTX 4090 D[/caption] GeForce RTX 4090 D (Dragon) will retain the same AD102 GPU as the original model According to insider MEGAsizeGPU, it will be AD102, but in a different version. If the RTX 4090 initially used the AD102-300, which was then replaced by the AD102-301, then the new product will receive the AD102-250. Typically Nvidia changes the index number depending on the number of changes. That is, AD102-250 will be noticeably different from AD102-300/301. Unfortunately, it is still unclear what exactly.  Let us remember that the sanctions mean that the GeForce RTX 4090 D (Dragon) should have performance below 4800 TOPS, but this is an aggregate figure for different modes. Nvidia can reduce this figure in various ways, including cutting the core and reducing frequencies.  Insider MEGAsizeGPU previously spoke about the launch of the RTX 4060 and RTX 4060 Ti before the announcement, as well as the timing of these models entering the market. 

[phonemantra-blog]

Video card cooling system - with a single centrifugal fan A typical GeForce RTX 4060 Ti from any manufacturer has two fans and occupies two slots in a PC case, but Colorful has prepared a special version of the accelerator - an especially thin one. The name of the model has not yet been announced, but there is a photo: it is visible that Colorful engineers have created a single-slot GeForce RTX 4060 Ti. [caption id="attachment_73634" align="aligncenter" width="780"] GeForce RTX 4060 Ti[/caption] Of course, a lot of things had to be redone here. And, in particular, the cooling system. Instead of two fans, one is installed - a centrifugal type. It is thin and probably noisy since the rotation speed must be high. [caption id="attachment_73635" align="aligncenter" width="780"] GeForce RTX 4060 Ti[/caption] The unique GeForce RTX 4060 Ti Colorful poses live. It only takes up one slot Power is supplied to a single eight-pin connector at the rear. The set of video outputs is standard: one HDMI and three DisplayPort. The length of the video card is 267 mm. [caption id="attachment_73636" align="aligncenter" width="450"] GeForce RTX 4060 Ti[/caption] Colorful has converted the GeForce RTX 4060 Ti into a single-slot version with 16 GB of GDDR6 memory. The video card is built on the AD106 GPU with 4352 CUDA cores. The memory bus width is 128 bits. Most likely, such an accelerator will not go on general sale: it is too specific, and performance in games may be far from maximum due to the poor cooling system. Most likely, the thin GeForce RTX 4060 Ti will be shipped directly to system integrators as a component in workstations.

[phonemantra-blog]

Video card cooling system - with a single centrifugal fan A typical GeForce RTX 4060 Ti from any manufacturer has two fans and occupies two slots in a PC case, but Colorful has prepared a special version of the accelerator - an especially thin one. The name of the model has not yet been announced, but there is a photo: it is visible that Colorful engineers have created a single-slot GeForce RTX 4060 Ti. [caption id="attachment_73634" align="aligncenter" width="780"] GeForce RTX 4060 Ti[/caption] Of course, a lot of things had to be redone here. And, in particular, the cooling system. Instead of two fans, one is installed - a centrifugal type. It is thin and probably noisy since the rotation speed must be high. [caption id="attachment_73635" align="aligncenter" width="780"] GeForce RTX 4060 Ti[/caption] The unique GeForce RTX 4060 Ti Colorful poses live. It only takes up one slot Power is supplied to a single eight-pin connector at the rear. The set of video outputs is standard: one HDMI and three DisplayPort. The length of the video card is 267 mm. [caption id="attachment_73636" align="aligncenter" width="450"] GeForce RTX 4060 Ti[/caption] Colorful has converted the GeForce RTX 4060 Ti into a single-slot version with 16 GB of GDDR6 memory. The video card is built on the AD106 GPU with 4352 CUDA cores. The memory bus width is 128 bits. Most likely, such an accelerator will not go on general sale: it is too specific, and performance in games may be far from maximum due to the poor cooling system. Most likely, the thin GeForce RTX 4060 Ti will be shipped directly to system integrators as a component in workstations.

[phonemantra-blog]

Looks like the ban will just take effect later The other day it became known that new US sanctions imply, among other things,  a ban on the supply of GeForce RTX 4090 video cards to China due to their high performance. Yesterday the information was clarified: supplies, as it turned out, are not prohibited, but production within China itself is prohibited. Now new data has appeared. [caption id="attachment_73560" align="aligncenter" width="780"] GeForce RTX 4090[/caption] GeForce RTX 4090 was removed from the list of sanctioned products for China The thing is that the initial information about adding the RTX 4090 to the sanctions list did not come from any third-party sources, but directly from the government website SEC.gov. There, the RTX 4090 was listed along with the A100, A800, H100, H800, L40 and L40S. However, in the current version of the document, after the update, the L40 and RTX 4090 were removed from the list. Against this background, one might think that the US government has changed its mind about the ban on the RTX 4090, but not everything is so simple. According to the Benchlife resource, the ban on the A100, A800, H100, H800, L40, and L40S comes into force immediately from the moment of publication, but the ban on the RTX 4090 and L40 will come into force on November 16. This information was reported to the source by several video card manufacturers at once, and the source speaks specifically about the ban on exports to China, and not just on production.

[phonemantra-blog]

So far with a very low share Valve has published the latest Steam hardware statistics. For the first time in the statistics, the desktop video card GeForce RTX 4060 appeared. Now they occupy 0.23%. For comparison, the mobile RTX 4060 Laptop already has 1.63%, but these adapters entered the market much earlier. [caption id="attachment_51940" align="aligncenter" width="780"] GeForce RTX 4060[/caption] GeForce RTX 4060 still buy. Video cards first appeared in Steam statistics Formally, the GTX 1650 remains the leader with a share of 5.11%, but, as we have said more than once, the RTX 3060 is the most popular video card, because before the RTX 30 line, Steam statistics combined mobile and desktop adapters of the same name into one position. That is, 5.11% of the GTX 1650 is the total share of the mobile and desktop GTX 1650. Whereas the desktop RTX 3060 has a share of 4.87%, and the mobile has another 4.29%. That is, in total, we will soon talk about 10% of all Steam gamers. If we talk about CPU, AMD's share continues to grow and now stands at 33.43%. The most popular are six-core processors with a share of 33.05%. At the same time, the shares of CPUs with a large number of cores are also actively growing.

[phonemantra-blog]

There's an RTX 4060, 4060 Ti and 4070 Colorful has introduced Ultra Z series video cards, which are distinguished by hidden power connectors.   [caption id="attachment_65221" align="aligncenter" width="780"] RTX 4060[/caption] The series includes the GeForce RTX 4060, RTX 4060 Ti, and RTX 4070 models. All of them are quite large (310 mm), but they use these dimensions not only to increase the CO radiators but also to extend the power cables into these same radiators. In addition, all this is covered with a lid, so the connector is truly invisible.  Colorful presented such models as part of the Ultra Z series In this case, the length of the cards also allows you to lay the cable in such a way as not to bend it near the connector itself.   The new items have not yet gone on sale, so it is not yet clear whether you will have to overpay for such a feature.