Best Eight-Core CPU Battle: AMD Ryzen 7 3800X vs Intel Core i7-9700K

[Ncogneto]

Depends on the overclock utilized. AOC and PBO are good, but manual overclock will do better in gaming if you disable SMT and are able to overclock 8 cores 8 threads to 4.5Ghz+. I know my 3800X is a good overclocker but I can hit 4.5Ghz all core SMT enabled @ 1.41V, SMT disabled I can do 4.575Ghz. Every 3800X I have worked with has hit 4.4Ghz all core with ranging voltages between 1.3 - 1.34V on average (LLC set to max). My bios is 1.0.0.3ABBA which for my X470 board has the best overclocking ability - from my experience 1.0.0.3ABBA is the best bios for X470 and X570 does the best with 1.0.0.4 (patch B). My drivers are all straight from AMD (not Asus) and are the latest available. I would assume using a X570 Godlike ($$$) they would be running the latest bios and drivers but who knows.

Not if they are simply writing a new article utilizing old test data compiled at or around the release date the the 3000 series CPU's and x570 motherboards. This is more common than you think.

Your running your voltages a bit higher than I care to, FWIW. But I agree, from my experience and what I have heard others say, 4.4 all core is pretty common on the 3800x with voltages in the 1.35 range.

 

[drivinfast247]

Gaming Performance Avg FPS (Geomean) Entire Test Suite 9700k OC 144.2 divided by 3800x PBO 127.0 =13.5%
99th Percentile FPS Entire Suite 9700k OC 102.9/3800x PBO 87.2=18%

If you aren't willing to take advantage of the easily available overclocks that can be achieved at safe temps, then you obviously don't really care about performance, but rather about appearances. Temps are not an issue with standard overclocks on Intel with a decent CL cooler.

If you are a gamer and can't come up with the cash for a 9700k and a closed loop water cooler then get the cooler and drop down to a 9600k, OC and you will still get 6.4% more average FPS gaming performance than any current AMD processor.

At 1440p or higher resolution?

 

[Redneck5439]

Not if they are simply writing a new article utilizing old test data compiled at or around the release date the the 3000 series CPU's and x570 motherboards. This is more common than you think.

Your running your voltages a bit higher than I care to, FWIW. But I agree, from my experience and what I have heard others say, 4.4 all core is pretty common on the 3800x with voltages in the 1.35 range.

Oh, I do agree, the voltages are too high. I never recommend above 1.34V for a 24/7 overclock and always recommend using a voltage offset so that isn't a static (non-stop) voltage pounding the chip even at idle. With my offset at idle I use ~ 0.5V. My max overclock I use 1.41V and can hit 4.5Ghz SMT on or 4.575Ghz SMT off, however my 24/7 overclock is 4.4Ghz @ 1.3V with an offset (much better for 24/7 use). I only use my max overclock for benchmarking, and some light tasks like gaming. All my overclocks are tested and Prime 95 stable, but I don't typically use it for rendering / editing / conversion or any other CPU heavy task, opting instead to use my 4.4Ghz settings. I do whatever possible to avoid damage / degradation.

I wouldn't be surprised if some of the test data was old / reused from Ryzen 3000 launch. New processors being sold now have better binning than at launch and typically perform better, new bios updates have increased performance as well as new chipset drivers... What I find really funny though is they say they are comparing the 9700K to the 3800X because of price but then throw on a very expensive custom loop and say its still in the same price range... The Intel setup would be easily be ~ $300 more with that custom loop. They should put a $40 air cooler on the 9700K and see how far it will overclock, that would be fair vs AMD stock cooling.

 

[Redneck5439]

At 1440p or higher resolution?

Why would you want to know about that??? Everyone knows you buy expensive high end Intel processors, the best premium custom loop cooling to get the best overclock possible, an overclocked RTX 2080Ti, an expensive 4K gaming monitor and play every game at 720p - 1080p. That's how you get the very best gaming experience... What is this 1440p for?? Who would want to game at 4K??😛

All joking aside though, I fully understand that the RTX 2080Ti of today will the be 3070 of tomorrow so you test at 1080p with the best GPU forcing the CPU to be the limiting factor. It also has to be said though that anyone buying high end processors, a 2080Ti, high end expensive 4K display isn't going to be playing at 1080p. At minimum they will be playing at 1440p, realistically they will be playing at 4K. At 1440p the difference between to Ryzen processors and Intel processors drops to around 4% and at 4K the difference is basically within the margin of error. Now some may say yea but I'm future proofed... well there is no such thing as future proofed period end of story. For example AMD may release some new technology that when you run a high end AMD GPU with a high end AMD CPU you get a boost in performance that outperforms next gen Nvidia and is the new standard for high end gaming... Now while the chances are very remote, stranger things have happened and no one knows what the future brings. Its much better to live in and game in the present, and that means that if you are going to buy expensive top of the line monitor, GPU and CPU you are looking at gaming at 1440p and 4K and there is very little difference in performance between Intel and AMD processors at these resolutions.

 

[superbrett2000]

Nice timing; I'm getting ready for a build in the next month or two. I've always been an Intel guy but this go around I'm planning on the 3700. The 3800 already destroys the i7 in the value proposition, and I'm unlikely to notice major real world performance differences by going with the 3700 over the 3800.

 

[Darkbreeze]

This would have been a lot more interesting had they used the 3700x, because it's unlikely that many people are flocking to buy the 3800x for 40 bucks more considering there is a 40w difference yet only a 100mhz difference in maximum boost speed and likely no actual difference at all once you factor in the increase in thermal conditions from the additional 40w TDP.

Which brings me to the second thing they should have done, which is to include an aftermarket cooler since we know that all these Ryzen processors tend to consistently boost higher on more cores with better cooling AND the fact that the Intel CPU had to have an aftermarket cooler anyhow. I don't think anybody buying a 3700x or 3800x is particularly stressed out about the idea of spending an extra 40 bucks on a CPU cooler that will likely net them another 200-300mhz peak boost, which might in fact be enough to either reduce or close the gap when it comes to the gaming performance comparison.

It just doesn't feel like a very realistic comparison when you've used a CPU that is not being commonly purchased and not putting an aftermarket cooler on a CPU that is commonly getting one in the wild.

 

[PaulAlcorn]

This would have been a lot more interesting had they used the 3700x, because it's unlikely that many people are flocking to buy the 3800x for 40 bucks more considering there is a 40w difference yet only a 100mhz difference in maximum boost speed and likely no actual difference at all once you factor in the increase in thermal conditions from the additional 40w TDP.

Which brings me to the second thing they should have done, which is to include an aftermarket cooler since we know that all these Ryzen processors tend to consistently boost higher on more cores with better cooling AND the fact that the Intel CPU had to have an aftermarket cooler anyhow. I don't think anybody buying a 3700x or 3800x is particularly stressed out about the idea of spending an extra 40 bucks on a CPU cooler that will likely net them another 200-300mhz peak boost, which might in fact be enough to either reduce or close the gap when it comes to the gaming performance comparison.

It just doesn't feel like a very realistic comparison when you've used a CPU that is not being commonly purchased and not putting an aftermarket cooler on a CPU that is commonly getting one in the wild.

We used a corsair h115i for both stock and OC on all processors, except where listed in the charts as otherwise.

 

[Ncogneto]

This would have been a lot more interesting had they used the 3700x, because it's unlikely that many people are flocking to buy the 3800x for 40 bucks more considering there is a 40w difference yet only a 100mhz difference in maximum boost speed and likely no actual difference at all once you factor in the increase in thermal conditions from the additional 40w TDP.

Which brings me to the second thing they should have done, which is to include an aftermarket cooler since we know that all these Ryzen processors tend to consistently boost higher on more cores with better cooling AND the fact that the Intel CPU had to have an aftermarket cooler anyhow. I don't think anybody buying a 3700x or 3800x is particularly stressed out about the idea of spending an extra 40 bucks on a CPU cooler that will likely net them another 200-300mhz peak boost, which might in fact be enough to either reduce or close the gap when it comes to the gaming performance comparison.

It just doesn't feel like a very realistic comparison when you've used a CPU that is not being commonly purchased and not putting an aftermarket cooler on a CPU that is commonly getting one in the wild.

this is a pretty typical OC on the 3800x
the 3700x will not do this on all cores.
As to if that is worth an extra $40.00, depends on your workload.
It is however quite easy to pick this processor up for $320 if you watch for sales.

Right now the I7 9700k is $60.00 more, and you still have to buy a cooler. And if you buy a $30.00 cooler, your not going to get the OC listed in this article. Take that extra $100 and buy yourself some good Samsung B-die memory for the Ryzen.
View: https://youtu.be/D428KLctpxU

 

[Darkbreeze]

We used a corsair h115i for both stock and OC on all processors, except where listed in the charts as otherwise.

Very confusing since in the test configuration chart there are hardware specifications for a number of products not even part of the comparison AND additionally there are references to:

Corsair H115i - Stock and OC cooler except where noted in charts
Custom Loop, EKWB Supremacy EVO waterblock, Dual-720mm radiators (HEDT only - n/a here)
AMD Wraith Spire, as noted


Except that, it's NOT noted anywhere as far as what is used with what and when. All in all, a very confusing read with no enough clarity and specificity of the information, and the inclusion of information that isn't relevant to the comparison which was the whole point of the article. Showing memory information for other CPU models, for example, isn't something that seems like it ought to be included in a comparison of CPUs when those CPUs aren't even part of the comparison.

Sorry, I think the article could be cleaned up a bit to offer a lot more clarity and include only the details that relate to those compared products, and add a bit more information such as, for example, whether the gaming performance on the Ryzen CPU was taken from results while it was overclocked manually, or overclocked with PBO, or not overclocked at all, and what cooler was used in each result. Too many question marks for me.

 

[CountMike]

Why only 3200MHz RAM when both CPUs can run much faster memory although Ryzen profits more from faster RAM ?
Another plus for 3800x is that it can run just as good on x470 and even good b450 MBs lowering system price by another $100 - $200 which can be put to good use by adding that money to better GPU and/or more/faster RAM, larger NVMe etc,

 

[Gurg]

What I find as interesting as we have yet another TH review with absolutely no data on the test setup. What cooler was the Intel CPU running? The AMD CPU?

Also fails to get into the memory settings, etc. Fine tuning a Ryzen CPU is a little different than your standard Intel "crank up the voltage and lets see what she can do" CPU".

I can get my 3800x to 4.4 ghz all core on 1.35v easy with decent cooling. All while running an FCLK of DD4 3800 (1 to 1). Not trying to say all 3800x CPU's can do this, but it is not unheard of. First thing you do with this CPU is pair it with some decent low latency ram and at a minimum run it at DDR4 3600 (FLCK 1:1).

Maybe the tester did this? No idea, again, no test system data.

The data for this review was from the July 2019 3600x review. Look back at it for more details.

https://www.tomshardware.com/reviews/amd-ryzen-5-3600x-review,6245-2.html

 

[Gurg]

Why only 3200MHz RAM when both CPUs can run much faster memory although Ryzen profits more from faster RAM ?
Another plus for 3800x is that it can run just as good on x470 and even good b450 MBs lowering system price by another $100 - $200 which can be put to good use by adding that money to better GPU and/or more/faster RAM, larger NVMe etc,

There was an article last summer about the 3000 series performance being around 3% lower on the older motherboards. Supposedly it will run but not as well.

 

[CountMike]

There was an article last summer about the 3000 series performance being around 3% lower on the older motherboards. Supposedly it will run but not as well.

x370 MBs had problem with T topology for memory but x470 is fine and had problems with RAM faster than 2933MHz on first gen Ryzen. No such thing with x470 chipset MBs.
My 3700x on Asus Prime x470 pro (AGESA 1004b) boosts to 4.4GHz and RAM is running at 3600MHz Cl16.
Only thing that may hamper 3800x on b450 MBs would be VRM that is generally lesser on those boards but nothing that good cooling can't fix.
About 3rd gen Ryzen OC, there may be some silicon lottery ivolved but if I can run 100% stable OC at 4.35GHz and less than 1.4v, I don't see why 3800x couldn't do 100MHz more. 3800x is just better binned 3700x, nothing special.

 

[calken]

I'm assuming from the review and the pricing that the $30 cooler is capable of of getting the 9700K to 5.1Ghz and keeping it there in the gaming and office benchmarks?

 

[Nick_C]

Oh dear - the Intel CPU was tested under water (as it doesn't come with a cooler) whereas the AMD one was tested with the stock air cooler. Not really a like-for-like comparison.

The lack of stock cooler with the 9700K also changes the value proposition - as the Corsair H115i costs around $140 at the moment. If that $140 is added to the c.$380 for the 9700K then the $520 cost of CPU + cooler means that a 3900X is a better cost-for-cost comparator as it comes with a stock cooler too.

 

[PaulAlcorn]

Oh dear - the Intel CPU was tested under water (as it doesn't come with a cooler) whereas the AMD one was tested with the stock air cooler. Not really a like-for-like comparison.

The lack of stock cooler with the 9700K also changes the value proposition - as the Corsair H115i costs around $140 at the moment. If that $140 is added to the c.$380 for the 9700K then the $520 cost of CPU + cooler means that a 3900X is a better cost-for-cost comparator as it comes with a stock cooler too.

The Intel vs. AMD processors were not tested with different cooling solutions, except where specifically called out in the charts.

 

[PaulAlcorn]

Why only 3200MHz RAM when both CPUs can run much faster memory although Ryzen profits more from faster RAM ?
Another plus for 3800x is that it can run just as good on x470 and even good b450 MBs lowering system price by another $100 - $200 which can be put to good use by adding that money to better GPU and/or more/faster RAM, larger NVMe etc,

3200 for stock operation, as that is the AMD spec, and 3600 for OC, as listed in the table.

 

[PaulAlcorn]

Very confusing since in the test configuration chart there are hardware specifications for a number of products not even part of the comparison AND additionally there are references to:

Corsair H115i - Stock and OC cooler except where noted in charts
Custom Loop, EKWB Supremacy EVO waterblock, Dual-720mm radiators (HEDT only - n/a here)
AMD Wraith Spire, as noted


Except that, it's NOT noted anywhere as far as what is used with what and when. All in all, a very confusing read with no enough clarity and specificity of the information, and the inclusion of information that isn't relevant to the comparison which was the whole point of the article. Showing memory information for other CPU models, for example, isn't something that seems like it ought to be included in a comparison of CPUs when those CPUs aren't even part of the comparison.

Sorry, I think the article could be cleaned up a bit to offer a lot more clarity and include only the details that relate to those compared products, and add a bit more information such as, for example, whether the gaming performance on the Ryzen CPU was taken from results while it was overclocked manually, or overclocked with PBO, or not overclocked at all, and what cooler was used in each result. Too many question marks for me.

The Corsair cooler was used for all test results, expert where explicitly called out in the test charts. In other words, if it isn't listed as otherwise, we used the corsair. We list the memory for all tested chips because it varies by chip generation/type. The charts say 'Ryzen 7 3800X' for stock testing, or 'Ryzen 7 3800X PBO' to indicate that the processor was tested with PBO.

 

[ern88]

Another win for AMD is that if you have an AM4 board. You can upgrade to a Ryzen 3 for the most part. Intel on the other hand. You have to buy a new motherboard. I'm glad I bought a Ryzen 7 3700x. Good value for the money. I got it on sale too lol

 

[Ncogneto]

3200 for stock operation, as that is the AMD spec, and 3600 for OC, as listed in the table.

At what timings?

* for anyone else just now reading my earlier comments, the test setup was not listed in the article initially, that has since been corrected, albeit somewhat confusing.

 

[Ncogneto]

3200 for stock operation, as that is the AMD spec, and 3600 for OC, as listed in the table.

Not seeing an entry in the charts for all core OC, although they claim to have done that? Can you also confirm that the 3800x was first release silicon?

Why are you using the wrong pricing information in your article? Suggested retail and not current pricing?

What bios revision and AGESA?

If this was from a July article, improvements have been made.

 

[PaulAlcorn]

At what timings?

* for anyone else just now reading my earlier comments, the test setup was not listed in the article initially, that has since been corrected, albeit somewhat confusing.

DDR4-3600 @ 14-14-14-34

 

[PaulAlcorn]

Not seeing an entry in the charts for all core OC, although they claim to have done that? Can you also confirm that the 3800x was first release silicon?

Why are you using the wrong pricing information in your article? Suggested retail and not current pricing?

We purchased the chip at retail soon after the launch. AMD did not sample this model to press.

Our overclocking efforts are detailed here: https://www.tomshardware.com/reviews/amd-ryzen-7-3800x-review,6226-2.html We used the PBO configuration for this article, as that provides the best balance of performance in both single- and multi-threaded workloads.

Unfortunately the retail pricing fluctuates considerably, so we're kinda just throwing a dart at the wall in that case. It could change in an hour, a few minutes, or a few days: There isn't a constant. The only constant is the suggested retail pricing, unfortunately. It isn't a perfect system, but that is what the retailers suggest, so there's that.

 

[Nick_C]

The Intel vs. AMD processors were not tested with different cooling solutions, except where specifically called out in the charts.

Given that the only CPU where cooling solution is mentioned in the charts is the R5 3600X, can it then be taken that the Corsair H115i was used in all other quoted tests?

That's still a $140 additional cost when the AMD CPU comes with a free, decent, cooler - which certainly changes the cost proposition.

Interestingly, from this review: https://www.tomshardware.com/uk/reviews/amd-ryzen-5-3600x-review,6245-3.html , it seems clear that the 3600X uses less power than the 3800X which, in turn, uses significantly less power than the 9700K.

If stock air cooling had been used, with an air cooler procured for the Intel CPU comparable to that provided with the AMD CPU, then the results might have been somewhat different.

 

[PaulAlcorn]

Given that the only CPU where cooling solution is mentioned in the charts is the R5 3600X, can it then be taken that the Corsair H115i was used in all other quoted tests?

That's still a $140 additional cost when the AMD CPU comes with a free, decent, cooler - which certainly changes the cost proposition.

Interestingly, from this review: https://www.tomshardware.com/uk/reviews/amd-ryzen-5-3600x-review,6245-3.html , it seems clear that the 3600X uses less power than the 3800X which, in turn, uses significantly less power than the 9700K.

If stock air cooling had been used, with an air cooler procured for the Intel CPU comparable to that provided with the AMD CPU, then the results might have been somewhat different.

The stock cooler for the 3800X is designed for 124W, whereas the PPT (the peak power consumption possible) is 142W, as mentioned in the original review:

"Like the other Ryzen 7 and 9 chips, the 3800X comes with the capable Wraith Prism RGB cooler which is rated to dissipate up to 124W if you crank the fans up to high. Given the 3800X's maximum 142W PPT measurement, that means, at least on paper, that the Wraith Prism might come up a tad shy of dissipating the full heat output of the 3800X at stock settings. "

However, upon testing we found very little meaningful variation between the Corsair and the stock cooler. We went with the Corsair results to provide a level playing ground. You can see similar variation in the charts with the 3600X with the Spire vs the Corsair charted out - The difference between Spire and the Corsair, with PBO active (peak consumption), amounts to a 0.2 fps variance in 99th percentiles, while looking at average fps yields a 0.8 fps variation. AMD's coolers are superb.

You are correct, the bundled cooler provides an exceptional value, as we covered in the article under the "cooling" section. Hence, AMD wins that category.