Ryzen smoother than i7 in gaming?

[dexT]

In this blind test generally they state all systems perform very similar and feel good but all 3 people had observed that the Intel systems had hitch and stutter at times. Wendell is legit so i think there could be something here. Do Ryzen adopters here notice the same?

 

 

[Drak3]

Review numbers have been largely agreeing. Better minimums, and oddly, some reviewers even noticed that AMD cards did better than Nvidia cards on Ryzen.

All in all, I'd rather have an R5 than an i7 if it means less frame dips. Especially if the marriage of a Ryzen CPU + AMD GPU is more consistent than my 1080 and 5930K are.

[dexT]

5 minutes ago, Drak3 said:

Review numbers have been largely agreeing. Better minimums, and oddly, some reviewers even noticed that AMD cards did better than Nvidia cards on Ryzen.

All in all, I'd rather have an R5 than an i7 if it means less frame dips. Especially if the marriage of a Ryzen CPU + AMD GPU is more consistent than my 1080 and 5930K are.

Extreme dips could be what they're describing but an i7 and 1080 shouldn't. 

[Moress]

I've seen a lot of better FPS minimums from Ryzen, so yeah I would be tempted to think the video is real

[SCHISCHKA]

the stuttering is the extra evil that intel puts into every processor

[Drak3]

Just now, dexT said:

Extreme dips could be what they're describing but an i7 and 1080 shouldn't. 

My minimums on my 1080 are in the 20's, sometimes mid thirties. Minimums on my 1070 when it was on the same system was also in the twenties and mid thirties, both on my 5930K and 4460. All 1080p, and oddly enough, the 1080 sees the same rough minimums @ 1440p. Then there is my RX470, mid thirties are the lowest 1080p minimums I've seen so far, but averages and high frame counts are sitting are 1/2 that of the 1080 @ 1440p (I didn't bother with avg. and high 1080p numbers for it, I got a 1440p monitor at the same time), so the 1080 is both pulling significantly more avg FPS and doing it at a higher resolution.

AVG for the 470 in Heaven: 70-75 FPS @ 1080p

AVG for the 1080: 120-150FPS @ 1440p

Same trend for every game I've tested. AMD's card is more consistent, I always know what to expect. With Nvidia's, I might get flawless performance today and the day after tomorrow, but tomorrow can be a stutter fest shortly after every cut scene and loading screen.

[WereCat]

7 minutes ago, SCHISCHKA said:

the stuttering is the extra evil that intel puts into every processor

Actually, I have noticed that in many benchmarks the 7700k has way worse 1% and 0.1% lows than the 4790k. It does have only better highs and averages.

[HelplmChoking]

Unless you're really hunting for that extra fps or two, which you wouldn't notice anyway, it really, really doesn't matter. For gaming, each are at the very top end and it's like comparing a ferrari to a lamborghini, one may be a little faster but they're both insanely fast and you'll be happy each way.

[dexT]

21 minutes ago, Drak3 said:

My minimums on my 1080 are in the 20's, sometimes mid thirties. Minimums on my 1070 when it was on the same system was also in the twenties and mid thirties, both on my 5930K and 4460. All 1080p, and oddly enough, the 1080 sees the same rough minimums @ 1440p. Then there is my RX470, mid thirties are the lowest 1080p minimums I've seen so far, but averages and high frame counts are sitting are 1/2 that of the 1080 @ 1440p (I didn't bother with avg. and high 1080p numbers for it, I got a 1440p monitor at the same time), so the 1080 is both pulling significantly more avg FPS and doing it at a higher resolution.

AVG for the 470 in Heaven: 70-75 FPS @ 1080p

AVG for the 1080: 120-150FPS @ 1440p

Same trend for every game I've tested. AMD's card is more consistent, I always know what to expect. With Nvidia's, I might get flawless performance today and the day after tomorrow, but tomorrow can be a stutter fest shortly after every cut scene and loading screen.

That sounds awful, at 1080p?

 

@Rangaman42 I find it interesting. I'm a gamer but enjoy benching as much.

[Drak3]

Just now, dexT said:

That sounds awful, at 1080p?

Thankfully, it's not that large of a problem when I'm gaming. I play games like Oneshot more than AAA titles where framerates are absolutely important. In Heaven, the problem is lessened after a full run through.

[JoostinOnline]

I get so annoyed when people say "i7" like it's a single CPU.  That could be like 50 different CPU's, how am I supposed to know which one you're talking about?  That's like giving directions to someone, but only telling them the street it's on.

[dexT]

Just now, Drak3 said:

Thankfully, it's not that large of a problem when I'm gaming. I play games like Oneshot more than AAA titles where framerates are absolutely important. In Heaven, the problem is lessened after a full run through.

Damn dude my 980 is thug at 1080p and I never see those minimums unless I'm getting low on vram(GTA 5 advanced graphics maxed). 

[Rhett Quigley]

im getting a 1080

[dexT]

15 minutes ago, JoostinOnline said:

I get so annoyed when people say "i7" like it's a single CPU.  That could be like 50 different CPU's, how am I supposed to know which one you're talking about?  That's like giving directions to someone, but only telling them the street it's on.

5960X and 7700K. Multiple SKUs of i7 so I used the term because they're the best ones for gaming.

[Drak3]

Just now, dexT said:

Damn dude my 980 is thug at 1080p and I never see those minimums unless I'm getting low on vram(GTA 5 advanced graphics maxed). 

I don't spend much time at the minimums, otherwise I would have returned both my 1070 and 1080 for either more 470's, or get 2 480's and put the 470 in the node 202 build. No Xfire, each is dedicated to a monitor

[dexT]

12 minutes ago, Drak3 said:

I don't spend much time at the minimums, otherwise I would have returned both my 1070 and 1080 for either more 470's, or get 2 480's and put the 470 in the node 202 build. No Xfire, each is dedicated to a monitor

Do you get what they were describing?

[Not_Sean]

I really wish reviews would completely change to just Min and Average FPS. I couldn't care if my system hits 200fps if its dipping below 60fps at points. 

 

I've seen the same in alot of the benchmarks I've seen where the Ryzen lineup thrashed the intels in Minimum fps, I say thrashed as even >20% thats a huge gap when we remember price difference, lets also factor in that people were tearing R7 apart as "Bad for Gaming" with like >5% worse than 7770k So sorry intel Defenders, but we'll hold you to the same standard and with min fps gaps this big.. Ryzen is winning hugely. 

[Drak3]

9 minutes ago, dexT said:

Do you get what they were describing?

I do, but most reviewers that include numbers for X99 chips see the same trend. Even if average is slightly lower, minimum is higher, relative to the Kaby lake quads.

 

Like I said, I get lows in the 20's on the 1070 and 1080, I don't spend much time there, but I do get those low numbers. It doesn't surprise me that chips like the 7700K does. Minimums don't always mean much, especially without knowing where the GPU spends that time (either jumping from min to max constantly, or hanging close to the avg most of the time), and if that dip is related to weird asset loading (which would be hard to measure without a graphic overlay synced to the output).

[othertomperson]

15 minutes ago, Not_Sean said:

I really wish reviews would completely change to just Min and Average FPS. I couldn't care if my system hits 200fps if its dipping below 60fps at points. 

GamersNexus are quite good for this. They give 0.1% and 1% lows, and they include error bars on their results so you can see the variance and consistency from run to run.

 

AMD fanboys sent them death threats because they refused AMD's request to benchmark Ryzen only in heavily GPU-bound tests, but they got some good data as a result, and the value of the 1600X shone through as a result.

[sgloux3470]

I don't have an i7 to compare it to but even at 4K where I've always been largely GPU bound on my FX 8370 I'm noticing a pretty large improvement.  I haven't finished benchmarking but here's some numbers for comparison.  I tried to benchmark mostly in Last Titan Standing since it is the most demanding game mode.  The results are very similar, but the game feels a lot better with Ryzen.  I'm actually amazed to see that the data is within margin of error because they feel completely different.  Before seeing the data I would have sworn the Ryzen system was running at a perfect 60 FPS the entire time while I could definitely notice the minor dips and stutters on the 8370.  It was sitll a solid experience, but Ryzen has felt perfect. 

 

 

 

FX 8370 @ 4.7 GHz  DDR3 1600 MHz

 Average FPS 58.12

1% Percentile 55.90

0.1% Percentile 50.32

 

Preliminary Results

R5 1600 @ 3.9 GHz  DDR4 2933 MHz

Average FPS: 58.27 1% Percentile 53 0.1% Percentile 50