AMD Radeon RX 5700 XT

[Space Lynx]

AMD Radeon RX 5700 XT review

A compelling blend of 7nm tech, RDNA architecture updates, and a last minute price drop.

www.pcgamer.com

@W1zzard we were talking about testing the cards at lower settings, well PCGamer did the work for you, Medium setting preset. oddly enough at 1080p the results are interesting, showing the 2070 super destroying the 5700 xt, versus just leaving settings on ultra. very interesting really. some of those ultra settings are just unrealistic performance hindering, skewing the more realistic results common gamers might see. i do wish they did high settings though not medium or ultra. bit of a shame.

 

[HTC]

Navi is looking really good imo, New arch, lets see how software adapts and of course OS/driver/API optimisations.
also

really cool.

This release reminds me of Zen1 launch a bit rough around edges but you can see the perf is there, and I suppose the 3870 in some ways.

There's no way 5700XT is faster than 2080Ti for gaming + streaming when that card is so much slower for gaming. Something else is going on and tests should be conducted in order to figure out what exactly that something is.

 

[Space Lynx]

There's no way 5700XT is faster than 2080Ti for gaming + streaming when that card is so much slower for gaming. Something else is going on and tests should be conducted in order to figure out what exactly that something is.

it's faster in Battlefield V and I think DOOM 2016. those are the only two and only when Vulkan/DX12 enabled. but yeah 2 games out of hundreds isn't saying much.

but yeah if more developers would go to Vulkan, the 5700 XT would beat the 2080 ti in those games. sadly that probably will never happen. or at least it is very slow going.

 

[A.Stables]

There's no way 5700XT is faster than 2080Ti for gaming + streaming when that card is so much slower for gaming. Something else is going on and tests should be conducted in order to figure out what exactly that something is.

Did you actually watch the video?

 

[ValenOne]

I was actually serious.
On the driver side, Navi uses the same old GCN driver with some minor tweaks, so the core driver should be very "mature". In fact, probably the most "mature" they've ever released. There isn't anything in the driver to optimize for Navi as it uses the same ISA. There might be improvements in the driver over time, but this will be general improvements unrelated to Navi.

NAVI has single clock cycle wave32 and two clock cycle wave64 compute length support while GCN has four clock cycle wave64.

 

[HD64G]

 

[HwGeek]

Radeon RX 5700 XT with an EKWB RX 5700 easily overclocked to 2.1 GHz [Max MSI AB limit]

igor´sLAB | PC & Components | Reviews & News

www.tomshw.de

EK-Vector Blocks Engineered for AMD Navi GPUs - ekwb.com

EK® Water Blocks, the leading premium computer liquid cooling gear manufacturer, is releasing two EK-Vector Radeon™ RX 5700 +XT water blocks that are compatible with reference design AMD® Radeon™ RX 5700 and 5700 XT graphics cards. This efficient and elegant-looking cooling will allow your...

www.ekwb.com

 

[ratirt]

I'd like to see the performance of the 5700XT when paired with a ryzen 3700 or 3800. Or any ryzen actually. Wonder if there's any difference in performance compared to previous ryzen generations CPUs.

 

[Xuper]

Did you test about feature : Anti-lag ?

 

[W1zzard]

I'd like to see the performance of the 5700XT when paired with a ryzen 3700 or 3800. Or any ryzen actually. Wonder if there's any difference in performance compared to previous ryzen generations CPUs.

check the pcie scaling article

 

[Midland Dog]

i want a comparison between the 57000 and the 580/590 at the same frequency

 

[efikkan]

I'm wondering how much of this is caused be the driver? If the driver has much to do with the power states of the card and it is not working correctly, it may cause a 'snowball' effect. Inefficient management could mean higher voltages, high voltages means more power and heat, which in turn, generally causes further increases in voltage as the card tries to stabilize.<snip> Worse case is that the quality of the first batch of 7nm dies are good, but not great. It's probably more likely that AMDs cooler, along with the early drivers, are just not up to scratch.

Managing power states is more a firmware thing than a driver thing.
While uncalibrated power states certainly can cause unnecessary power consumption, this should be one of the first things they calibrate. And considering how delayed Navi has been, I don't see any reason why they left this uncalibrated.

 

[ratirt]

check the pcie scaling article

Some of the games are better on intel and some not. Depending on the game. The most significant difference is in F1. 1080p on Ryzen shows 50FPS more and 1440p 20fps more when you pair 5700XT with new Ryzen. Also Rage 2 shows some differences. Other games are basically same fps +- 0-3% difference. Civilization VI showed some noticeable differences as well.
I've compared some of the benchmarks from TPU RX 5700XT review and PCI-e 4.0 review. I need to go through all the games and see what the difference is.

 

[W1zzard]

Some of the games are better on intel and some not. Depending on the game. The most significant difference is in F1. 1080p on Ryzen shows 50FPS more and 1440p 20fps more when you pair 5700XT with new Ryzen. Also Rage 2 shows some differences. Other games are basically same fps +- 0-3% difference. Civilization VI showed some noticeable differences as well.
I've compared some of the benchmarks from TPU RX 5700XT review and PCI-e 4.0 review. I will need to go through all the games and see what the difference is.

I'm not sure if F1 is comparable, because the game tends to reset settings on CPU change, which might explain why the FPS difference is so huge. You should be able to identify those cases by looking at 4K, which should really be within 1%, due to being completely GPU limited

 

[Space Lynx]

I'm not sure if F1 is comparable, because the game tends to reset settings on CPU change, which might explain why the FPS difference is so huge. You should be able to identify those cases by looking at 4K, which should really be within 1%, due to being completely GPU limited

I don't get why F1 is even tested, pretty sure less than 10% of this community plays that. Probably best to just pick best selling 15 games in last 2 years by volume/profit. I don't know honestly, /shrug I am sure you and other sites have your reasons

 

[ratirt]

I'm not sure if F1 is comparable, because the game tends to reset settings on CPU change, which might explain why the FPS difference is so huge. You should be able to identify those cases by looking at 4K, which should really be within 1%, due to being completely GPU limited

Well I didn't mention 4k since the percentage of difference was very small. Either way I will look over all of the benchmarks to see if there is any difference. Thanks

I don't get why F1 is even tested, pretty sure less than 10% of this community plays that. Probably best to just pick best selling 15 games in last 2 years by volume/profit. I don't know honestly, /shrug I am sure you and other sites have your reasons

Benchmarking was never about to test newest games but to test variety of different games kinds which may have (I assume) different impact on the GPU itself and processor.
Newest games are mostly glitchy and are not mature enough for optimal or correct FPS number. They need some tweaking.

 

[RichF]

Review puts lack of CrossFire in the negatives category, then writes:

"Abolishing CrossFire support might seem sad at first, but I think it's actually a good thing."

Which is it? Is it a positive or is it a negative?

 

[W1zzard]

Review puts lack of CrossFire in the negatives category, then writes:

"Abolishing CrossFire support might seem sad at first, but I think it's actually a good thing."

Which is it? Is it a positive or is it a negative?

Good question, what do you think? "think" is the whole goal of my reviews, I'm not trying to decide for you

 

[Space Lynx]

Good question, what do you think? "think" is the whole goal of my reviews, I'm not trying to decide for you

I think it is positive, its too few games anyway, and now developers can consistently put more effort into single gpu performance instead of trying to do to much at once and wasting precious resources during game development.

 

[RichF]

Good question, what do you think? "think" is the whole goal of my reviews, I'm not trying to decide for you

Then the logical course of action is to place the CrossFire issue into a neutral visual category or to list it in both the positive and negative visual categories — while not stating that it's solely a positive in the summary text.

As the review appears currently, it is contradicting itself.

 

[W1zzard]

Then the logical course of action is to place the CrossFire issue into a neutral visual category and/or to list it in both the positive and negative visual categories — while not stating that it's a positive in the summary text.

As the review appears currently, it is contradicting itself.

we don't have a neutral category, i usually leave out things that are in-between, but many people just scan through the pro/con list, that's why i included it. not sure if it even matters, as long as people are aware. this is similar to the "no dvi output" discussion in the past, or "no analog vga"

 

[RichF]

we don't have a neutral category, i usually leave out things that are in-between, but many people just scan through the pro/con list, that's why i included it. not sure if it even matters, as long as people are aware. this is similar to the "no dvi output" discussion in the past, or "no analog vga"

I realize that there is no neutral category which is why I gave two options that would resolve the contradiction. Only one of them involves creating a separate neutral visual category.

 

[HD64G]

Just re-posting some of my previous ones before my final verdict on the Navi GPUs

I hope most here have already undrestood what the slides show. +25% IPC means that the comparison between 5700 vs Vega64 without core clocks mentioned gives a +25% advantage in performance to Navi, being 50% more efficient at the same time. To make things simple, if we put those numbers on the diagrams from the latest @W1zzard GPU test 5700 sits exactly between the 2070 and the Radeon7 and consumes about 200W. If price is good, that will be a great product. As for Real time tracing, not any GPU has the power yet to allow that feature maxed out to run constantly over 60FPS in big resolution. So, for 2020 the big Navi might be the one for that.

1) If AMD's performance slides are real, we are talking about a GPU having the same performance as Radeon VII and being sold for much less, while beating 2070 and costing less also. When nVidia lowers the prices, AMD will have the ability to lower theirs too. Competition lowers the prices, not wishes. Be real people.
2) Reference cooler is the best for cases that don't have great airflow and it is to protect the product from damage due to the pc onwer ignorance on this aspect of pc building Custom models wil come soon.

So, the price after the last minute reduction is good (1080Ti/2070S/2080/Radeon VII performance for much less money) but not great for a Polaris successor. Efficiency could be better but it is top-notch taking advantage of the 7nm process and the new arch. Maybe some smaller dies will do even better when they come out later. All in all a great product to bring more competition in the GPU market so kudos to AMD. Custom 5700XTs will shine and allow for big oc. And imho, the big Navi can challenge for the top performance tier if it comes out soon enough.

 

[Xuper]

here we go .AMD ruined the whole RX 5700 series.....

https://twitter.com/i/web/status/1148113823137533952

67'c , That's damn good.

 

[Turmania]

TPU used to test World of Warcraft before. Haven't seen them tested for a while.is there a specific reason for it? I know that game is hard to test as it is affected by many things but it is the one I play the most...