Originally posted by vick
View Post
Announcement
Collapse
No announcement yet.
Intel vs. AMD Performance-Per-Watt On Ubuntu 14.04 Linux
Collapse
X
-
-
Originally posted by guido12 View PostWill these graphs be updated or should I do what tuke81 did for the other tests?
Comment
-
Originally posted by H.F.P View PostThe confusion comes from the term "power consumption". Many people think "power consumption" to be equal to "energy consumption". In fact, the energy consumption is the integral of power consumption over time. Therefore your complaint is more about the fact that you would like to see a conclusion about "energy consumption". On the other hand, the problem with energy consumption comes with workloads that usually run "forever" or at least not with a fixed amount of time e.g., playing a computer game. For the latter, power consumption can be treated equivalent to energy consumption.
Anyhow, it can't be repeated often enough: consuming a little bit more power over a much shorter duration still holds a lower energy consumption.
Comment
-
Originally posted by guido12 View PostWill these graphs be updated or should I do what tuke81 did for the other tests?
I actually came here to post a comment about the graph fail, but was gratified to see that everybody is already all over it.
Seriously, how could those graphs in the article ever have been published? They're so obviously, obviously wrong. You don't even have to think - they leap out and slap you around the head with a wet fish.
Finally, (physicist here) I'd like to see two values plotted:
Performance per Watt = benchmark result / (average power * time / time) = benchmark result / average power over task (units of per Watt)
Energy needed per task = average power * time taken (units of Joules)
Of course, the performance per Watt formula assumes higher is better, so for those benchmarks where lower is better you'd need to use 1/result or something for the performance.
Comment
-
Originally posted by baffledmollusc View Post
Finally, (physicist here) I'd like to see two values plotted:
Performance per Watt = benchmark result / (average power * time / time) = benchmark result / average power over task (units of per Watt)
Energy needed per task = average power * time taken (units of Joules)
Of course, the performance per Watt formula assumes higher is better, so for those benchmarks where lower is better you'd need to use 1/result or something for the performance.
I'm not sure why people keep suggesting all sorts of strange metrics. The only important metric here for power efficiency is the total energy required to complete the task. Of course, it is also useful to know how quickly the task is completed, and that is already provided by the "Time to Compile" benchmark.
Bottom line is that this type of benchmark should include two values for each CPU: energy required to complete task, and time required to complete task.Last edited by jwilliams; 18 January 2014, 08:00 PM.
Comment
-
Originally posted by devius View PostI wouldn't hold my breath. This isn't the first time Michael posts obviously wrong performance-per-watt results, people complained, provided the correct formulas to use and yet here's another bad set of results. And I'm sure it's not the last either, so if you'd like to actually post the correct results I'm also interested in thatOriginally posted by baffledmollusc View PostI doubt it. I've never seen Michael fix an article, no matter how egregious the error.
I actually came here to post a comment about the graph fail, but was gratified to see that everybody is already all over it.
Seriously, how could those graphs in the article ever have been published? They're so obviously, obviously wrong. You don't even have to think - they leap out and slap you around the head with a wet fish.
Finally, (physicist here) I'd like to see two values plotted:
Performance per Watt = benchmark result / (average power * time / time) = benchmark result / average power over task (units of per Watt)
Energy needed per task = average power * time taken (units of Joules)
Of course, the performance per Watt formula assumes higher is better, so for those benchmarks where lower is better you'd need to use 1/result or something for the performance.
I guess I'll go slap something in Octave because I'm interested in this article's subject for the new mini-ITX PC I plan on building. I haven't completely decided on Kaveri or Haswell yet.
Comment
-
Originally posted by droste View Postreally? :-D
Say you get a benchmark score of 80 FLOPS, iterations, whatever. Say it took 30 seconds, and the CPU used an average power of 50W over that time. So
Performance per Watt = benchmark result / (average power * time / time)
= 80 FLOPS / (50W * 30s / 30s)
= 80 FLOPS / 50W
= 1.6 FLOPS / W
= 1.6 FLOPS per Watt
The reason I wrote it that way is average power * time = energy, and to get a power you have to divide by time taken.
Although now I look at it, I'm prepared to agree that while correct, it's not the most straightforward approach. I think the issue is I always think in terms of energy, and convert that to whatever later.
Comment
-
Originally posted by baffledmollusc View PostI doubt it. I've never seen Michael fix an article, no matter how egregious the error.
Test signature
Comment
Comment