According to the latest data, Nvidia has again postponed the release of the GeForce RTX 4090 graphics card to October. These were supposedly the original plans of the company, then the announcement was moved a month earlier, but after Nvidia revised the plans and returned everything to the original version. The reason lies in the large stocks of GeForce RTX 30 cards that need to be sold first. By the way, a significant reduction in official prices should help the company in this.
A new generation may be announced as early as September, but the first card, and this will be the RTX 4090, will be released in October. Unfortunately, recent data also suggests that all other video cards of the new line will not be released at all until 2023. Earlier, recall, it was said that the RTX 4080 will be released in October, and the RTX 4070 in December. Such a move would give Nvidia even more time to sell out the current generation.
Also, sources attribute the RTX 4090 TDP model to 450W, not 600W. Apparently, 600 W will still be consumed by the RTX 4090 Ti, which will be released sometime later. So the RTX 4090 will be on par with the RTX 3090 Ti.
Almost $5,000 for an Nvidia video card with a 256-bit bus, but 32 GB of memory. RTX 5000 Ada appeared in retail
Nvidia has launched the RTX 5000 Ada graphics adapter. This is a professional solution that should not be confused with the regular RTX 5000, since the latter is a previous generation model.
The RTX 5000 Ada has been available for about a month now, but previously it could only be purchased as part of ready-made workstations. Now it has appeared in retail, although so far, apparently, only in China.
There they ask for about $4,800 for a new product. This is about half the price of the RTX 6000 Ada, but the parameters of the younger model are simpler. The card offers an AD102 GPU, but with only 12,800 active CUDA cores. At the same time, the new product has 32 GB of memory, but a 256-bit bus. Performance (FP32) is 65.3 TFLOPS with a TDP of only 250 W.
Neither in Blender nor in V-Ray can the new Core i9-14900K outperform the “old” Ryzen 9 7950X
The results of testing the Core i9-14900K processor in tasks that can actually load modern multi-core processors have appeared on the Internet.
In Blender, the new product scores almost 560 points, which is 6% more than the Core i9-13900KS. This is good considering that these are technically identical CPUs. The new product is already 8% ahead of its direct predecessor, the Core i9-13900K. However, the Ryzen 9 7950X is still a little faster.
In V-Ray, the new Intel flagship showed a result of 27,790 points, ahead of the Core i9-13900KS by 5% and its predecessor by 8%. But here again the AMD flagship is ahead.
Of course, it was clear in advance that Raptor Lake Refresh would only be a very minor update (except for the Core i7, which will get more cores), but now we can verify this not only through the results from Geekbench.
The Ryzen 8000 and Ryzen 9000 processors will offer more cores than their predecessors. AMD roadmap describing Zen 5 and Zen 6 leaked online
An AMD roadmap has leaked online, describing a number of key features of the company’s future products. It was obtained by the author of the Moore’s Law Is Dead channel.
The document, among other things, describes the Zen 5 and Zen 6 architectures.
The first to hit the market, of course, will be Zen 5, codenamed Nirvana. The first CPUs based on it are expected in the first half of 2024. Processors with Zen 5 architecture will be produced using 3 and 4 nm technical processes and will bring an increase in IPC (the number of instructions executed per clock cycle) by 10-15%. It is also known that the data cache of Zen 5 will be increased to 48 KB, a new low-power core version will appear, and Zen 5 will also bring a 16-core cluster. Probably, only Zen 5c cores will be combined with 16 pieces, but this still means that the total number of cores in the new processors will increase and finally exceed the 16 pieces mark that we have become accustomed to since the Ryzen 3000.
Zen 6, codenamed Morpheus, is expected to move to 3nm and 2nm process technology and will be released sometime in 2025. The IPC indicator will increase by 10%, that is, not as much as in recent years. But the architecture will bring support for FP16 computing to accelerate AI/ML algorithms and a new memory profiler. Moreover, processor clusters will be able to have 32 cores each, which means that we will again see an increase in the number of cores in the CPU as a whole.
Moore’s Law Is Dead profiled many AMD products before they were released. In particular, the Radeon RX 7800 and Radeon RX 7700 video cards several months before the announcement.
“Some of our clients didn’t believe the data until they put it to the test.” HBM3 Gen2 memory is already being tested by Nvidia
Micron has already started shipping HBM3 Gen2 memory to Nvidia, and the testing results seem to be very impressive. The...
China’s SMIC is only four years behind TSMC and Samsung, although US sanctions should keep the gap to 10 years
The Chinese company SMIC has recently become much more famous due to the fact that, as it turns out, it...
Changan CS35 Plus Super Edition presented: two pedals, two screens and 160 hp. for 11 thousand dollars
The Changan CS35 Plus crossover is already inexpensive, but with the Super Edition it has become even more interesting: the...
The newest Geely Monjaro L with a consumption of only 4.3 l/100 km is already being filmed for advertising: photo
The Geely Monjaro L is preparing to enter the Chinese market: the car was previously certified by the local Ministry...
News3 days ago
The Firefox browser now has a built-in page translator that works even without the Internet
News6 days ago
BelAZ presented super-new products: a 130-ton hydrogen dump truck and a 200-ton hydraulic excavator
Electric Cars3 days ago
“We have a huge problem,” Volkswagen and Audi plants around the world shut down
News4 days ago
110 inches, 8K, 240 Hz and more than 40 thousand backlight zones. Hisense TV UX unveiled – one of the most advanced TVs in the world