The averages are not the same as the median values, I think this is where some of the problem comes from. The plots have quartiles with the boundaries shown as lines. The line showing the median value for the Ampre system is near the middle of the plot, but the median value for the AMD plot is far over to the right end of that plot, suggesting that many of the results were in a narrow range just above that value. This would skew the total average energy consumption way up, so we would see the difference shown in average Joules per run. This is probably not a good type of plot for this type of data, a scatter plot or line chart may be better.

Heres more: * first chart has a n of 3. (Mythbusters' rocket car had n>3) * the jouls you reference have candle charts showing way too much variance to make any conclusions.

Never-mind that these are all reduced to absurd levels, or biased.

My favorite was some site crapping on a SSD that only managed 3GiB/s for 100GiB of data, then dropped to 500meg or something. But, they didn't mention data transferred at all. Just speed vs time. Obviously pushing for that higher kickback on the ssd that costs 4x as much and uses 8x the power.

It's a what you can buy today vs. what you can buy today comparison. Ampere chose to use N5 even though N3 was available and they are paying for that decision.

>since nowadays there's no guarantee that a process shrink will get you significantly cheaper transistors

That is because all cutting-edge chips use TSMC.

No competition means price per transistor can stay consistent or even rise, which is one part of why most modern CPUs and GPUs have price/performance ratios that are the same or worse than their previous-generation counterparts.

>can you really have an apples-to-Apples comparison if the chips being compared aren't on the same process node?

Of course not, but that isn't going to stop people from doing it, nor is it going to stop people from going "x86 is dead" when comparing last-gen-node AMD processors to CPUs only Apple can use (conveniently forgetting that Qualcomm's products underperform at the same process node).

Shouldn't all the credits go to TSMC anyway? I mean coming up with an architecture for a GPU is no small feat, but it's nothing compared to building a fab with the capabilities of TSMC's.

I own a 5950x, that 130 number sounds extremly wrong for idle. I'd recommand checking settings in bios. I'm getting around 7-20 depending on which number I read in Ryzen Master, so adding in the gpu idle of around 30-40, I'm around half of 130 for the entire system! I'll also add low idle power would make scence for a server owner, so I could see AMD priorizing it..

That idle is unusually high, you should check for your cpu governor (if the host is Linux, use AMD state with balanced, otherwise refer to Windows power settings). If you talk about system idle (instead of CPU idle) then a dGPU can also have a significant impact.

130?

My systems normally idle at 30-60.

There is a huge difference between Mainboard chips etc.

None of my systems idle at 130. Not even my 2x 4090 system for ml

Oracle buys the first year of production then they become available to the public later. AmpereOne should hit NewEgg around the time it becomes completely obsolete.

I hope so, but it clearly isn't the present, unless you're aware of a RV processor in this league that I don't yet know about?

Future can be 6 days from now or 6 centuries from now. This statement is useless without specific details.

Sure buddy. Just one little thing please tell me where you found a RV64GCV system with comparable throughput as well as throughput per watt instead of a ~100MHz in-order dual-issue toy core that doesn't exist outside FPGAs (and emulation).