Chester (lamchester.delete@this.gmail.com) on November 23, 2020 1:27 pm wrote:
> > > IPC at stock clocks will naturally favor lower clocked CPUs, because memory latency
> > > will be lower in terms of clocks. Because of that, I think measuring IPC at identical
> > > clocks is better if you're trying to directly compare architecture.
> > >
> > > IPC scaling with clock speed should be tested separately, but they'd need performance
> > > counters to account for DVFS and sadly most review sites don't do that.
> >
> > This is wrong and debunked often in this forum, but it wasn't my point, so I will ignore it.
> >
> > Take the 1800X as example. Stock RAM is 2667MHz and in Zen µarch MEMCLK and DFICLK
> > are tied, which implies when you overclock the RAM, you are overclocking IF. This means
> > independently of the core clock the CPU is no more in stock settings. In this review
> > memory and IF were overclocked by 20% increasing IPC over stock configuration.
>
> Yes, it would increase IPC over the stock config. It also lets the author separate CPU
> perf from memory perf as much as possible.

No. Precisely that MEMCLK and DFICLK are tied in the Zen µarch, means that CPU and memory are tied and once you overclock the RAM you are overclocking the CPU.

> But anyway, I think 4 GHz, DDR4-3200 is a fine
> comparison point. Testing IPC at stock wasn't the author's goal to begin with.

Irrelevantly to which was his goal, he is comparing stock chips to non-stock chips. Moreover, he chose a narrow and rather biased selection of applications (basically rendering, rendering, rendering...) and the gaming testing was not a proper CPU test.

> > > > and not providing full CPU tests in the gaming measurements.
> > >
> > > What's not 'full' about CPU tests in the gaming measurements?
> >
> > 1080p is not a full CPU test.
>
> What kind of test would you like to see? Higher resolutions push
> more load to the GPU, and tend to minimize CPU differences.

If you pay attention to my message, I already answered your question. The answer is "full CPU test".