Synthetics

Moving on, we have our synthetic performance testing, taking a look at geometry throughput, pixel throughput, memory compression, and more.

Synthetic: TessMark - Image Set 4 - 64x Tessellation

Given the significance of cutting a quarter of Navi 10’s GDDR6 memory bus, I was very curious to see what that would do for synthetic performance. But for better or worse, this has created more questions than it has solved.

The numbers listed below are accurate, in as much as these are the results I get when testing these cards. Whether they are correct, however, is another matter.

The problem, in short, is that due to AMD’s very aggressive power savings/idling implementation for their Navi 10 cards, I have been unable to get these cards to run at their full memory clockspeeds when executing the the Beyond3D Suite benchmark suite. The GPU clocks regularly pass 1600MHz like they should, however AMD’s telemetry is reporting that memory clocks are rarely hitting 7Gbps, let alone 12Gbps+. As a result, we end up with results like the pixel test below, where the RX 5600 XT is beating the RX 5700, an otherwise impossible outcome.

As best as I can tell, this issue has been going on since the launch of the Radeon RX 5700 series back in July, but it’s only now that I’ve noticed it, in large part due to the RX 5600 XT cards being slightly less aggressive in their idling. In other words, those cards are boosting to higher memory clockspeeds more often, putting them ahead of the RX 5700 and bringing the clocking issue front and center.

I’m still working on a proper fix for the issue, but for now the results with Navi 10 cards should be taken with a large grain of salt. The benchmark itself is still fine, but AMD’s aggressive power management (and lack of an easy means to disable it) is kneecapping AMD’s performance in these benchmarks.

Synthetic: Beyond3D Suite - Pixel Fillrate

Synthetic: Beyond3D Suite - Integer Texture Fillrate (INT8)

Synthetic: Beyond3D Suite - Floating Point Texture Fillrate (FP32)

Synthetic: Beyond3D Suite - INT8 Buffer Compression

Synthetic: Beyond3D Suite - FP32 Buffer Compression

Compute Power, Temperatures, & Noise
Comments Locked

202 Comments

View All Comments

  • Qasar - Thursday, January 23, 2020 - link

    " nvidia drivers are near flawless " ha !!!!!!!!!!!! that's a good one. bias much cmdmonkey ??? as Korguz mentioned, you must have been very lucky, as for me, with both radeon and geforce, i have had issues with both over the years, but, that doesnt change my view on either, i buy what i can afford that gives the performance i am willing to pay for, and so far, that looks to be amd, and thats partly because i am sick of nvidia charging WAY to much for a video card, they must like pricing their cards so only the rich, or those with no families and mortgages, can buy them. i hope amd can do the same with the video cards, as they have with Zen. would be nice to see nvidia have to drop their prices by 800 or more like intel has had to do.
  • Spunjji - Thursday, January 23, 2020 - link

    Nvidia drivers have - in admittedly rare cases - actually destroyed cards. In more regular cases, they have bugs with setting resolutions, bugs with installation, and just the general sort of issues you'd expect from a large driver for complex hardware. They're no more immune from that than anyone else.

    This "AMD drives are baaaaad" trope will probably never die, but it'll never really be true either.
  • cmdrmonkey - Thursday, January 23, 2020 - link

    For the record I'd actually love to see AMD produce good video cards again. Everything they've made in recent years has been trash. They make loud, hot, power guzzling, underperforming cards with bad drivers that rarely get updated. That's why they are down to 11% market share. Their cards suck.
  • Qasar - Thursday, January 23, 2020 - link

    oh look, an opinion from some that that looks like they hare heavily bias against amd for what ever reason. while their video cards in recent years hasnt been that good, but to be fair, there HAS been years, where nvidias were just as bad, same with their drivers. i guess he LOVES paying WAY to much for nvidia's products and wants to keep overpaying... is it safe to assume you also like over paying for intels current products too ?? nvidia is getting lazy, just like intel, and who knows, maybe they will follow in intels footsteps too.....
  • cmdrmonkey - Thursday, January 23, 2020 - link

    I actually hate overpaying and I hate nVidia having a monopoly. I want AMD to pull their heads out of their butts and stop producing video cards that are trash.
  • Qasar - Thursday, January 23, 2020 - link

    amds cards aren't trash, they are still pretty good for the price, and with nvidia charging way to much for theirs, amds cards are still quite usable, and at least most of them are affordable.
  • cmdrmonkey - Thursday, January 23, 2020 - link

    AMD cards are poo. And they aren't cheap enough that anyone would pick them over nVidia equivalents that run cooler and quieter and have better drivers.
  • Qasar - Thursday, January 23, 2020 - link

    your opinions are poo, and is just your personal OPINION, thats all.. there is nothing wrong with amds cards, just like there is nothing wrong with nvidias, some have issues with one, and not the other, and vice versa, as i said, i have issues with both. still better then paying the ridiculous prices that nvidia charges, as i also mentioned.. nvidia is getting lazy, been at the top too long, and it will bite them in the ass, maybe with the next release of RDNA2, or heck.. maybe it will come out of left field with intel and the Xe cards who knows.. IMO, your bias against amd shows.... and i hope you keep buying nivida and paying for those over priced cards....
  • flyingpants265 - Thursday, January 23, 2020 - link

    Hahaha, this post is embarrassing, you should stop posting on Anandtech.
  • 335 GT - Friday, January 31, 2020 - link

    Why did the 5700xt win videocard of the year on the every site I saw? Shoo troll.

Log in

Don't have an account? Sign up now