[Techspot] AMD Ryzen 5 1600 vs. Intel Core i7-7800X: 30 Game Battle! [Links Fixed - Updated]

[scannall]

Test are meaningless. But opinions about the hardware are worthless.

I often see people with their arms flapping around after seeing tests on particular hardware eager to shout that one brand sucks and the other is better. How many times have we seen this over past months/years, that initial reviews of hardware shown one brand having worse performance, and after some time, the second, acclaimed to be worthless hardware outpaced the "better" competitor?

It doesn't matter if it's computer components, automobiles or your minty fresh toothpaste. Some people get emotionally invested in the products they buy. Seeking some kind of validation on their wisdom.

 

[TheELF]

Yeah it seems that 7700k is the best in current State of the Art 720p low quality gaming. For whoever plays at this resolution.
How the picture will look like on 2018 software? Dont know but i dont think software from 2013-2016 will be predictive.

Click on the link and look at the numbers,the 7700k get's much better FPS then both the hexacores,intel's and AMD's, in 1080p ultra with the best GPU available in the world,they should run these tests in 720 low since there are no GPUs that can cope with modern games now in 1080/ultra but with 2018-19 new GPUs the lead of the 7700k will only get bigger and it would be the right thing to do to show this to the customer.

 

[Vaporizer]

Click on the link and look at the numbers,the 7700k get's much better FPS then both the hexacores,intel's and AMD's, in 1080p ultra with the best GPU available in the world,they should run these tests in 720 low since there are no GPUs that can cope with modern games now in 1080/ultra but with 2018-19 new GPUs the lead of the 7700k will only get bigger and it would be the right thing to do to show this to the customer.

It might become bigger in the Softwares tested with upcoming GFX Cards, however i do not know how relevant these Games will be in 18/19 for the average Gamer (e.g. Battlefield1 might be less relevant then). How it will look like in the Softwares to come in 18/19 no one knows and the data doesn not allow for any prediction.

 

[IEC]

You're looking at pretty much the worst case performance for AMD today. Plenty of game engines not optimized for their core topology, brand new architecture, less than maximum RAM speeds. With their much increased investment in developer support for AAA studios and low-hanging fruit for optimization, by the time Zen 2 rolls around I would expect a 2018 game test suite to show decisive improvements in gaming performance at low resolutions.

For a $200 processor to be having such great value out of the gate for gaming shows the potential of the Zen architecture.

 

[Anarchist Mae]

Click on the link and look at the numbers,the 7700k get's much better FPS then both the hexacores,intel's and AMD's, in 1080p ultra with the best GPU available in the world,they should run these tests in 720 low since there are no GPUs that can cope with modern games now in 1080/ultra but with 2018-19 new GPUs the lead of the 7700k will only get bigger and it would be the right thing to do to show this to the customer.

That 720p low gaming demonstrates future gaming performance hasn't been clearly correlated to the best of my (limited) knowledge. Has anyone actually tested this?

 

[moonbogg]

That 720p low gaming demonstrates future gaming performance hasn't been clearly correlated to the best of my (limited) knowledge. Has anyone actually tested this?

The best it can do is show how a current game might perform with a faster GPU from the future. It can't predict what games will come in the future and how they will handle CPU power. Its kind of pointless IMO because people don't wait around for an entire generation of GPU's to pass before buying a game. I don't see people testing BF3 at 1080p with a GTX 1080ti and proclaiming, "See! We were totally right about that 3770K being 6% faster than the 2600K". At the time it didn't matter because everyone was bottlenecked by a GTX 570 or something. No one bought a faster CPU back then hoping to get 200FPS in BF3 when the GTX 1080ti comes out in 5 years. Screw this lame back-to-the-future game testing methodology.

 

[Actaeon]

The Techspot review seems to be an outlier compared to other sites that has benchmarked SKL-X. No other site has results as poor for the 7800x as this one.

I think Digital Foundry has the best and more comprehensive gaming focused review of SKL-X so far.

1. They compared the 7740x, 7800x, 7820x, and 7900x to show what the extra cores get you.
2. They did an interesting 8-core comparison between the 7820x, 6900k, 5960x, and 1800X, all at 4.4ghz (except the Ryzen @ 4.0) to demonstrate IPC.
   
3. They compared stock and OC results of the 7740x (5.0), 7820x (4.8), 7900x (4.6g).
4. They also compared the 7900x w/ 2133mhz ram vs 3200mhz at stock and OC (4.6)

http://www.eurogamer.net/articles/d...view-core-i9-7900x-i7-7820x-i7-7800x-i7-7740x

 

[formulav8]

No other site has results as poor for the 7800x as this one.

Did they test 30 games?

 

[raghu78]

Did they test 30 games?

Very well said. No website tests such a huge number of games. The good thing about techspot is the game suite is so large it covers games of all types - games which are primarily driven by single thread performance, games which scale well to 8 threads or even more etc. So you get a good idea of how these CPUs play across a wide range of games. The summary game perf bar chart is very useful

https://www.techspot.com/review/1450-core-i7-vs-ryzen-5-hexa-core/page9.html

 

[Actaeon]

Did they test 30 games?

You don't need to test 30 games to see that the results don't quite match the other sites. Here are results of the same games from two other sites except they put the 7800x versus the 1800X instead of the 1600.

• https://nl.hardware.info/reviews/74...intel-core-i7-7800x7820x-vs-amd-ryzen-7-1800x
  • Hardware.info has the 7800x faster or equal to the 1800X at every game they tested, averaging out to 12.7% faster in gaming overall.
    • In GTA V, Techspot shows the 7800x as being 2-3% slower than a 1600 but hardware.info shows the 7800x as being 6 to 15% faster than the 1800X.
    • In BF1, Techspot shows the 7800x as being less than 1% slower than a 1600 but hardware.info shows the 7800x as being 17 to 31% faster than the 1800X.
    • In Doom, Techspot shows the 7800x as being 10% slower than the 1600, but hardware.info shows the 7800x being equal to the 1800X.
    • In RoTR, Techspot shows the 7800x as being 12% faster than the 1600 but hardware.info shows the 7800x being 16-18% faster than an 1800X
• http://www.eurogamer.net/articles/d...view-core-i9-7900x-i7-7820x-i7-7800x-i7-7740x
  • Digital Foundry showed that the 7800x was faster in every game they tested than a Ryzen 1800X OCed to 4.0.
    • In Farcry Primal, Techspot showed that the 7800x was 3% slower than a 1600 @ 4.0ghz, but Digital Foundry shows the 7800x being 15% faster than an 1800X @ 4.0ghz
    • In The Division, Techspot showed that 7800x was equal to a 1600x @ 4.0ghz, but Digital Foundry shows the 7800x being 5% faster than an 1800X @ 4.0ghz

 

[Zucker2k]

As soon as pictures emerged of the toasted 7800x (due most probably to misaligned pin), I knew those TechSpot results should have been canned. Also, it doesn't matter if you test 100 games, if you're not testing CPU demanding games (so you know where you stand with future gaming demands), why even bother overclocking the 7800x? The chip that benefitted the most from overclocking was the 1600x, indicating it's anemic stock performance. The delta between stock and overclocked 7800x on the other hand, was negligible - indicating it's powerful performance, even at stock.

 

[MarkPost]

Very well said. No website tests such a huge number of games. The good thing about techspot is the game suite is so large it covers games of all types - games which are primarily driven by single thread performance, games which scale well to 8 threads or even more etc. So you get a good idea of how these CPUs play across a wide range of games. The summary game perf bar chart is very useful

https://www.techspot.com/review/1450-core-i7-vs-ryzen-5-hexa-core/page9.html

Agreed

 

[DownTheSky]

You don't need to test 30 games to see that the results don't quite match the other sites. Here are results of the same games from two other sites except they put the 7800x versus the 1800X instead of the 1600.

• https://nl.hardware.info/reviews/74...intel-core-i7-7800x7820x-vs-amd-ryzen-7-1800x
  • Hardware.info has the 7800x faster or equal to the 1800X at every game they tested, averaging out to 12.7% faster in gaming overall.
    • In GTA V, Techspot shows the 7800x as being 2-3% slower than a 1600 but hardware.info shows the 7800x as being 6 to 15% faster than the 1800X.
    • In BF1, Techspot shows the 7800x as being less than 1% slower than a 1600 but hardware.info shows the 7800x as being 17 to 31% faster than the 1800X.
    • In Doom, Techspot shows the 7800x as being 10% slower than the 1600, but hardware.info shows the 7800x being equal to the 1800X.
    • In RoTR, Techspot shows the 7800x as being 12% faster than the 1600 but hardware.info shows the 7800x being 16-18% faster than an 1800X

They tested with memory @ 2666Mhz. That's why. While HU tested with mem @3200.

 

[coercitiv]

They tested with memory @ 2666Mhz. That's why. While HU tested with mem @3200.

And Digital Foundry had a BIOS version that enforced all-core turbo to max turbo.

Benchmarks here were taken with the standard all-core turbo disabled, but the processors still seemed to run at max turbo speeds regardless, giving a very close grouping, with only the 7800X falling a touch behind.

But then again, we all know these benchmarks should not be directly compared, at least not without careful examination of system specs and testing procedures.

 

[Anarchist Mae]

As soon as pictures emerged of the toasted 7800x (due most probably to misaligned pin), I knew those TechSpot results should have been canned. Also, it doesn't matter if you test 100 games, if you're not testing CPU demanding games (so you know where you stand with future gaming demands), why even bother overclocking the 7800x? The chip that benefitted the most from overclocking was the 1600x, indicating it's anemic stock performance. The delta between stock and overclocked 7800x on the other hand, was negligible - indicating it's powerful performance, even at stock.

This again? I asked earlier in this thread if anyone had demonstrated that this kind of testing has any predictive value. Nobody has so far. New games are just as likely to be GPU bottlenecked as current games, and we're now seeing developers take multi-threading more seriously.

 

[Zucker2k]

This again? I asked earlier in this thread if anyone had demonstrated that this kind of testing has any predictive value. Nobody has so far. New games are just as likely to be GPU bottlenecked as current games, and we're now seeing developers take multi-threading more seriously.

You don't think current, stronger cpus would be better suited to future games than weaker ones? Especially, taking your last sentence into consideration? What do you think will happen if faster gpus are released alongside more demanding multi-threaded games in the future?

 

[Anarchist Mae]

You don't think current, stronger cpus would be better suited to future games than weaker ones?

I'm questioning if running benchmarks of games at settings nobody would actually play those games at (720p low for example) has any merit for predicting future gaming performance.

Plenty of publications have done such benchmarking in the past, but I'm not aware of any doing the necessary follow up to demonstrate that it was actually useful and did predict what future gaming performance was.

Especially, taking your last sentence into consideration? What do you think will happen if faster gpus are released alongside more demanding multi-threaded games in the future?

I think that at the settings most people play at and depending on the type of game they will either be GPU or CPU bottlenecked.

It is certainly more useful to me to see currently available games benchmarked on settings I would be playing them at than it is to see currently available games benchmarked at settings I would not be playing them at in order to demonstrate what a publication thinks future performance will be even though they do not know and never check after the fact.

 

[TheELF]

Benchmarks are supposed to show you the absolute best that's possible,reviews are supposed to show you everyday usage.
Don't confuse the two and don't rebuke people that don't confuse them.

 

[Markfw]

Benchmarks are supposed to show you the absolute best that's possible,reviews are supposed to show you everyday usage.
Don't confuse the two and don't rebuke people that don't confuse them.

Thats your opinion. Do you have proof of what you say ? If not then don't harass those who do not believe as you,

 

[scannall]

You don't think current, stronger cpus would be better suited to future games than weaker ones? Especially, taking your last sentence into consideration? What do you think will happen if faster gpus are released alongside more demanding multi-threaded games in the future?

We will continue to be GPU bottlenecked at higher resolutions for a while. In 2020 or so when either pcie4 or pcie5 hits the mainstream that may chage then. But everyone will need a new CPU and motherboard to use it, regardless of team red or team blue.

 

[Zucker2k]

I'm questioning if running benchmarks of games at settings nobody would actually play those games at (720p low for example) has any merit for predicting future gaming performance.

Plenty of publications have done such benchmarking in the past, but I'm not aware of any doing the necessary follow up to demonstrate that it was actually useful and did predict what future gaming performance was.



I think that at the settings most people play at and depending on the type of game they will either be GPU or CPU bottlenecked.

It is certainly more useful to me to see currently available games benchmarked on settings I would be playing them at than it is to see currently available games benchmarked at settings I would not be playing them at in order to demonstrate what a publication thinks future performance will be even though they do not know and never check after the fact.

But this is not what I stated at all. What I said was, CPUs that handle today's cpu-demanding games best, will again do better in the future. Or you reckon the R5 1600x is going to outperform the R7 1800x in cpu-demanding games (both at 4ghz) in the future? Where are dual-cores now? Even AMD's quads are nowhere to be seen. The only reason Skylake and Kabylake have done so well is ipc and high clocks. But even they struggle when a well-threaded, cpu-demanding game is encountered. The 1600x is not a bad processor, but those wanting to game on Ryzen a bit longer should aim for the octocore, to make up for the 4Ghz hard limit. It'll serve them well in the future.

 

[Anarchist Mae]

But this is not what I stated at all. What I said was, CPUs that handle today's cpu-demanding games best, will again do better in the future. Or you reckon the R5 1600x is going to outperform the R7 1800x in cpu-demanding games (both at 4ghz) in the future? Where are dual-cores now? Even AMD's quads are nowhere to be seen. The only reason Skylake and Kabylake have done so well is ipc and high clocks. But even they struggle when a well-threaded, cpu-demanding game is encountered. The 1600x is not a bad processor, but those wanting to game on Ryzen a bit longer should aim for the octocore, to make up for the 4Ghz hard limit. It'll serve them well in the future.

It may not be what you stated, but the effect is the same. Introducing an artificial CPU bottleneck while benchmarking today may not predict future performance, and to the best of my understanding has never previously been used to successfully predict performance.

And at the end of the day, it still doesn't tell me as a consumer what kind of system I need to build to enjoy the games I want to be playing today.

 

[Zucker2k]

It may not be what you stated, but the effect is the same. Introducing an artificial CPU bottleneck while benchmarking today may not predict future performance, and to the best of my understanding has never previously been used to successfully predicted performance.

And at the end of the day, it still doesn't tell me as a consumer what kind of system I need to build to enjoy the games I want to be playing today.

What artificial bottleneck? Are there no cpu-demanding games today with in-built benchmarks?

 

[Crumpet]

Current situation allows us to clearly see who is bigoted and who is logical. Its astonishing how many bigoted people live among us. Much more than I thought.

It may not be what you stated, but the effect is the same. Introducing an artificial CPU bottleneck while benchmarking today may not predict future performance, and to the best of my understanding has never previously been used to successfully predict performance.

And at the end of the day, it still doesn't tell me as a consumer what kind of system I need to build to enjoy the games I want to be playing today.

Yup, I see the use for 720p low graphics testing to understand a cpu's absolute performance.

But what I want to see are real world scenarios and setups.

 

[Anarchist Mae]

What artificial bottleneck? Are there no cpu-demanding games today with in-built benchmarks?

You want them to only test CPU demanding games, not all games are CPU demanding, therefore you'd be artificially skewing the results of the 30 game test making it not representative of what general gaming is like. Note however, I'm not suggesting that individually benchmarking CPU demanding games is a bad thing, it's especially important if those games happen to be your main focus and should inform your purchase.