HW News - RAM & SSD Prices Will Fall, RTX 3090 Reports, TSMC Makes One Billion 7nm Chips
video description
Date: 2020-08-26
Related videos
Comments and reviews: 9
Tim
So when we do a CPU core and add an instruction to it, historically the power of x86 and our instruction set extensions have been that we made them available everywhere.
Except, of course, for all the chips that have certain instructions disabled due to product segmentation. That quote was a pretty bad lie.
reply
So when we do a CPU core and add an instruction to it, historically the power of x86 and our instruction set extensions have been that we made them available everywhere.
Except, of course, for all the chips that have certain instructions disabled due to product segmentation. That quote was a pretty bad lie.
reply
Charles
Quick question, if DRAM, and SSD prices go down why does the GPU space unaffected, this maybe related to the VRAM chips on the GPUs. Whenever we hear DRAM prices it's always RAM and SSD affected. Why is that the GPU is unaffected?
Don't know much about economics in general.
reply
Quick question, if DRAM, and SSD prices go down why does the GPU space unaffected, this maybe related to the VRAM chips on the GPUs. Whenever we hear DRAM prices it's always RAM and SSD affected. Why is that the GPU is unaffected?
Don't know much about economics in general.
reply
sadlerbw9
I, for one, would actually like to applaud IBM for using a numbering scheme for product generations that is NOT a nonsensical mess driven by unfathomable marketing directives. I will gladly take '9, 10, 11' over '9, 10, 20 and 16, 30' or 'XP, Vista, 7, 8, 10, 10, 10'
reply
I, for one, would actually like to applaud IBM for using a numbering scheme for product generations that is NOT a nonsensical mess driven by unfathomable marketing directives. I will gladly take '9, 10, 11' over '9, 10, 20 and 16, 30' or 'XP, Vista, 7, 8, 10, 10, 10'
reply
Jonas
AVX512 can be used for loop vectorization. This means if a lot of the same instructions are to be executed on a large set of data this can potentially bring very large performance benefits to applications, which are compiled with suppport for this instruction set.
reply
AVX512 can be used for loop vectorization. This means if a lot of the same instructions are to be executed on a large set of data this can potentially bring very large performance benefits to applications, which are compiled with suppport for this instruction set.
reply
rampage
what you mean ram prices will fall? i just bought new ram! tough 80 for 2x8gb 3600mhz ram aint to bad. the site i bought them from had the exact same ram sticks but slower for the same price hell one 2666mhz one was actualy more expensive than the 3600mhz one
reply
what you mean ram prices will fall? i just bought new ram! tough 80 for 2x8gb 3600mhz ram aint to bad. the site i bought them from had the exact same ram sticks but slower for the same price hell one 2666mhz one was actualy more expensive than the 3600mhz one
reply
soiTasTic
AVX-512 is great for a narrow set of workloads, of course the people who have workloads that profit from it love it. Coincidentally that's also one of the very few instances in which Intels Xeon still wins against Epyc, they would be crazy to give it up.
reply
AVX-512 is great for a narrow set of workloads, of course the people who have workloads that profit from it love it. Coincidentally that's also one of the very few instances in which Intels Xeon still wins against Epyc, they would be crazy to give it up.
reply
Siresly
Seemed like the right week, but looks like I picked the wrong week to buy RAM.
This particular kit dropped from 2300kr to 2000kr, it's historically lowest price. But apparently that won't last long.
reply
Seemed like the right week, but looks like I picked the wrong week to buy RAM.
This particular kit dropped from 2300kr to 2000kr, it's historically lowest price. But apparently that won't last long.
reply
Henrik
I actually used avx 512 (on a hpc) and performance for solving eigenvalue problems increases massively ! For usual consumers it makes no sense but in science it has a massive impact
reply
I actually used avx 512 (on a hpc) and performance for solving eigenvalue problems increases massively ! For usual consumers it makes no sense but in science it has a massive impact
reply
David
Well, I sold my 2080 Ti s, for damn near what I paid for them two years ago
Just hope 3090 launch isn t too far off cuz the thought of being stuck on PS4 and X1 makes me
reply
Well, I sold my 2080 Ti s, for damn near what I paid for them two years ago
Just hope 3090 launch isn t too far off cuz the thought of being stuck on PS4 and X1 makes me
reply
Add a review, comment
Other channel videos
