News China's first 6nm gaming GPU matches 13-year-old GTX 660 Ti in first Geekbench tests — Lisuan G100 surfaces with 32 CUs, 256MB VRAM, and 300 MHz cl...

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
I'm sure they'll catch up. But my point is the media, both Eastern and Western (WCCFTech does this too) credulously report every ridiculous claim like this, that the article immediately undercuts. But most people don't read articles, so they tend to believe the headline.
clickbait at its finest
 
somehow you proved XP is good enough, many people play old games online and stream video,
Of course XP is good enough to play old games! That's true, almost by definition!

What it lacks is stuff like support for NVMe SSDs and I'd guess it's probably not great at scheduling SMT on multi-core CPUs, either. Oh, and forget ReBAR or video drivers with DX12 or Vulkan support. Win 7 is probably the oldest OS that's half-decent at that sort of stuff.
 
I don't really laugh at this sort of stuff. We've seen inaugural GPU releases go this way before. Especially and most recently with Intel. Then the drivers get sorted out and it turns out quite decent. Intel never released a high end GPU but with their drivers largely figured out the early ars were up to par with 3060s and 3070s, where before they were 50-40%+ off from either. It's drivers. Coding. We're going to see some security advancement and competition in the tech industry. Namely on account of AI. Not just in the consumer market and enterprise but militarily.
 
  • Like
Reactions: ivan_vy
I don't really laugh at this sort of stuff. We've seen inaugural GPU releases go this way before. Especially and most recently with Intel. Then the drivers get sorted out and it turns out quite decent. Intel never released a high end GPU
But Intel had literally decades of experience with their iGPUs!! So, it wasn't surprising that the hardware was half-decent and they just needed time for their software to catch up.

I think it will take more iterations of hardware, in order for Chinese GPUs to find their stride. However, there is some older IP they might be leveraging, such as Imagination's or Vivante's.
 
I don't really laugh at this sort of stuff. We've seen inaugural GPU releases go this way before. Especially and most recently with Intel. Then the drivers get sorted out and it turns out quite decent. Intel never released a high end GPU but with their drivers largely figured out the early ars were up to par with 3060s and 3070s, where before they were 50-40%+ off from either. It's drivers. Coding. We're going to see some security advancement and competition in the tech industry. Namely on account of AI. Not just in the consumer market and enterprise but militarily.
Actually the reason of you rarely if ever see a PRC tech company succeed in real self made hardware is more of a political environment issue.

I am not saying about ideaology, it's what market/money they are trying to make.

In the west the money is made from real customers, be it individual consumers or big coporation customers, so they actually NEED to develop something working and improving it to gain money and be sustainable.

While in China, you always see the supreme leaders gave an order, allocate billions of funds to develop something they have zero or close to zero profession or infrastructure in to make it work in short time, and in order to get the praise and rewarded by hundreds of millions of funding, you need to please the leaders in a speed run, first come first serve.

So ended up all the budget and headlines comes from those scammers who have decades of experience in something else, like grazing pigs suddenly found a genius teenage girl or some experienced 70 years old lady leading the team and successfully made such and such, quite often being discovered some months later it is actually a polished and remarked actual old intel/nvidia chip in disguise.

Chinese have a lot of smart and talented ppl for sure, but in the environment, when you try to get funding for some real, decade long project of R&D into competing products, you will get outcompeted by those headline making propaganda favourite "miracle success" and ends up in where we are now.