I wonder why Nvidia doesn't establish it's own fab business. It might be a good fit for them.
ASML and Samsung seal deal on 2nm chips
71–80 of 97 posts
Re: ASML and Samsung seal deal on 2nm chips
#72Earlier quoted context omitted.
You joke but it does seem like we’re near the end of the line. We can go from 3 nm to 2 nm to 1 nm, but we can’t go negative nanometers. That’s it.
Next up is picometers, femtometers, attometers, zeptometers, yoctometers, rontometers, and quectometers (not to be confused with quesometers). They can just keep going until they hit the Planck length as they continue to divorce their marketing from reality.
And if we run out of prefixes we will add new ones.
Re: ASML and Samsung seal deal on 2nm chips
#73I wonder why Nvidia doesn't establish it's own fab business. It might be a good fit for them.
AFAIK the cost of producing semiconductors on a leading-edge node are so high you are forced to aggegate demand across seveal chip designers. Intel offering foundry services is evidence of this dynamic.
Edit: grammar.
Re: ASML and Samsung seal deal on 2nm chips
#74I wonder why Nvidia doesn't establish it's own fab business. It might be a good fit for them.
AFAIK the cost of producing semiconductors on a leading-edge node are so high you are forced to aggegate demand across seveal chip designers. Intel offering foundry services is evidence of this dynamic. Edit: grammar.
True, but aside the internal demand I don't see why can't they get external customers.
I think Samsung did the same.
Re: ASML and Samsung seal deal on 2nm chips
#75Earlier quoted context omitted.
Depends what you consider as part of chip production. If you consider new instructions and the related software as new chip production, then the performance improvements have continued quite dramatically. Single chip inference has gone 1000x over the last 10 years according to Bill Dally from Nvidia [1] at 10 minutes (probably not original source). Main gains according to the video have been different numeric types (…
So going from 65nm to 3nm only gave us 2.5x speed? If by speed we only understand frequency, then it might be true. But speed depends also on IPC and having more transistors means higher IPC.
Ah yes thanks! Makes sense. So, more transistors allow for the more complex instructions, which allow for more effective instructions per cycle (such as SIMD).
Re: ASML and Samsung seal deal on 2nm chips
#76Everything over the last years tells me we are very close to the end game in chip production. 1. power stopped scaling 2. clocks stopped - 4x over 20 years? 3. node names became fiction. 4. EUV now differentiates the top from everyone else. 5. IPC improvements are a trickle. 6. Tons of non-moores law ideas are here: A) GPUs B) chiplets C) stacked die D) specialized accelerators 7. Governments are now involved in divi…
We have a long way to go before we get to the end of the road in where we can go in chip design. The real sore spot moving forward is the grind it's going to be to improve price/performance ratios.
Increasingly I'm adjusting by being willing to spend more than I would previously consider reasonable on big ticket tech purchases, rather than waiting for prices to drop.
Re: ASML and Samsung seal deal on 2nm chips
#77Everything over the last years tells me we are very close to the end game in chip production. 1. power stopped scaling 2. clocks stopped - 4x over 20 years? 3. node names became fiction. 4. EUV now differentiates the top from everyone else. 5. IPC improvements are a trickle. 6. Tons of non-moores law ideas are here: A) GPUs B) chiplets C) stacked die D) specialized accelerators 7. Governments are now involved in divi…
I think us software types may be more sad about the end of the "free lunch" than the buying public. The 90s in particular had amazing performance increases, but you also needed to buy a new computer (which were more expensive then) every ~3 years, or upgrade roughly every 18 months if you wanted to stay near the top.
Anyway, my optimization skills will stay in demand :)
Re: ASML and Samsung seal deal on 2nm chips
#78Everything over the last years tells me we are very close to the end game in chip production. 1. power stopped scaling 2. clocks stopped - 4x over 20 years? 3. node names became fiction. 4. EUV now differentiates the top from everyone else. 5. IPC improvements are a trickle. 6. Tons of non-moores law ideas are here: A) GPUs B) chiplets C) stacked die D) specialized accelerators 7. Governments are now involved in divi…
Every time someone says we've hit peak microprocessing, we in fact have not hit peak microprocessing.
That can be true while approaching an asymptote. Unless there is a switch in technology (optical, different materials, whatever), I don't expect much for the time being. At least in CPUs and especially in single core performance. GPUs and other massively parallel devices are easier to improve with 3D techniques and such.
Re: ASML and Samsung seal deal on 2nm chips
#79Everything over the last years tells me we are very close to the end game in chip production. 1. power stopped scaling 2. clocks stopped - 4x over 20 years? 3. node names became fiction. 4. EUV now differentiates the top from everyone else. 5. IPC improvements are a trickle. 6. Tons of non-moores law ideas are here: A) GPUs B) chiplets C) stacked die D) specialized accelerators 7. Governments are now involved in divi…
We arent.
There are materials which could make cpus eg 1000x faster, but their life span would be e.g year or two and cost would be 10000x
>Tons of non-moores law ideas are here:
What?
Re: ASML and Samsung seal deal on 2nm chips
#80Is there a good comparison of how each fabs nanometer metric (aka version number) lines up?