Recent content by Stomx

This year both Intel and AMD will deliver their new chips made by their Oregon and Arizona latest fab 0.019874 nm technology. Versus previous year 0.02 nm technology chips this will bring 0.044% improvement in fps, TOPS, FP2 and FP4. Major BIOSes already made them compatible with all existing...

I already realized exactly this your dream a year ago and could not believe how good this ended up. The fate and progress of consumer processors do not bother me anymore except temporally for processors for ultraportables as I probably need to update also my laptop to match these high end...

When this misnomer of tech world and creature of finances Microsoft will finally go out of business? 99.8% of top supercomputers are finally just on one single OS - Linux. Most of cell phones are on Linux. More and more Chrome OS (Linux based) laptops which are twice cheaper than Microsoft...

Prices of 4TB NVME drives actually were as low as 3.4 cents per GB in November 2023. https://i.postimg.cc/nzKQWwNX/Screenshot-from-2023-11-26-00-01-47.png

Notices its rediculous kilomegagiga speeds in Steady Write Speed test ? Look at 2nd and 3rd screen. This drive will show off first 40 seconds its 6pack muscles and then blow into the mud like a stone running just as usual 4th gen drive. Avoid at any cost if you move large volumes because every...

moAr and cheAp )

Exactly. I have screenshot somewhere from Dec 2023 with the prices for 4TB NVMe SSD around $160 which you can not find anywhere even today

I had zero advertisements in Skype in the whole 20 years. And only probably just one scammer total. Skype is ideal for personal communications unlike for example regular emails interrupted whole day and night with myriads of other users, spammers and scammers. The only bad thing of Skype was...

Crash landing of fake nanometers everybody expecting eventually will happen? Imagine you just got the first sample of 18A with these parameters and walking to report it to CEO... Imagine what kind of night sleep CEO will have...

This is because poor NVIDIA can not even eat, all the time thinking about their users. And its Jensen Huang honestly believes that "...the more you buy the more you save"

I would be really impressed if they together with 38% increase of density *decreased* by 33% the number of stored bits per cell switching from TLC to MLC. That little sacrifice would bring a 1000% increase in endurance and reliability of flash long term. Yes not immediately but for sure 5-10...

Of course doubling the voltage to 24V would reduce the heating load by the factor of 4 and further quadrupling to 48V to the factor of 16 so that the whole power could be easily supplied by just a single (!) wire but here is one issue. I would worry about more than 24V for the GPU due to small...

Time to switch from 12V to 24V and the problem is solved for the next decade or two when the GPUs will consume 2.5 - 3 kiloWatt. By the way NVIDIA already switched to even higher voltages in its server GPUs

Damn, PCIe 5.0 is what i was waiting this card for (used for simulations)

More interesting numbers. If Elon Mask current AI cluster consisting of 100,000 H100 GPUs (which are not substantially slower than H200) would start training DeepSeek model at 9am in the morning it would finish it at approximately the end of the day at 7pm. And if the DeepSeek decided to...