AMD Ryzen 7 1800X CPU Review

Page 12 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
I tried this XFR-thing with a real good water cooling solution and also with iced water inside (<6°C). I saw no real differnce between both results.

 


Yeah I know, even Anandtech was more favourable than TH. I mean, really? Anandtech? Those guys were supposed to be THE Intel-Lovers' corner of the net and TomsHardware was supposed to be the respectable, unbiased site. How times have changed!
 


If the only review you've read is this one, I can't blame you for thinking that. Fortunately, this isn't the only review:
Hardware Canucks found it to be much better than a certain editor here:
http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/74814-amd-ryzen-7-1800x-performance-review.html
Even AnandTech was more favourable (how the hell did that happen?):
http://www.anandtech.com/show/11170/the-amd-zen-and-ryzen-7-review-a-deep-dive-on-1800x-1700x-and-1700
Forbes thinks that it's pretty damn good:
https://www.forbes.com/sites/patrickmoorhead/2017/03/02/amd-ryzen-desktop-amd-said-it-would-be-awesome-and-it-is/#234a594b2d48
Here we have a columnist from seekingalpha who actually came out and said that his own skeptical preview of Ryzen was dead-wrong:
http://seekingalpha.com/article/4051908-wrong-amds-ryzen

You might not be so disappointed after reading all of those.


 

Between you and me: TH and Anandtech are sister sites, both owned by Purch. And now? If Purch was paid by Intel, I have no idea what happened with Anandtech... 😀

You saw Lisa on Redit? She exactly said, what we (and a lot others) found. :)

Simply follow the expert discussions about latency problems, scheduler and the fact, that Windows 10 handles Ryzen worse than Windows 7 (and why). Nothing, what can't be solved with drivers, patches and improved micro-code. But we have to report the real status quo and not the cherry-picked best case. The launch was simply a bit too early for my taste.

 

In an interview with AMD about XFR, the AMD rep said that you can expect an extra 100MHz going from basic cooler to a better HSF and another 100MHz going to liquid cooling or high-end air. Every chip is different though, so maybe your sample didn't hit XFR's sensor thresholds for the next 100MHz step even with chilled water. Since it is "automatic overclocking" and people would not expect built-in overclocking to cause stability issues, XFR needs to stick to the safe side and could be excessively conservative.

Then again, with the number of reviews having very little overclocking success using conventional cooling, there does not appear to be much overclocking headroom in the first run of Ryzen chips. Ryzen under LN2 cooling yields some surprising results though.
 
I tried two CPUs and saw no difference between an cheap air cooler and real cold water. XFR works, but a bit limited.
I was thinking, it was designed for normal utilization and not only for LN2 games. I was on AMDs presentation in Sonoma to hear more about XFR. 😉
 


Well, I don't know about you, but I am most certainly NOT used to seeing AMD doing something like this. With the exception of this review that manages to keep Intel in the lead in nearly EVERYTHING, you'll find that Hardware Canucks and AnandTech are far more fair. And let's face it, for TomsHardware to be MORE biased against AMD than AnandTech means that something is DEFINITELY wrong here! I smell a rat and it's name is Intel. It looks like Charlie Demerjian was actually correct. Intel did manage to influence one of the biggest review sites on the net. I can honestly not believe that the owner of Tom's Hardware isn't embarrassed by this.

 
I'm not paid by anyone and I'm totally independend since years. And: Charlie is the best example for paid content. His statements are mostly not better than Trumps twitter posts. 😀

The most of European major sites were even harsher than we. And again: hear and see, what Lisa said on Reddit. And you will see, what is right and what was propaganda or simply wishful thinking. The things around Ryzen must be improved (patches, drivers, micro-code) to give this architecture a real chance. But without some findings and not cherry-picked benchmarks this will never be. Take a look at Intel. Without a reason they will never return from vacation-mode. I'm sure, reviews like ours are a good motivation to do this Ryzen thing to the end. The launch was simply a bit too early :)

 

Every review that I have seen shows Ryzen being quite competitive if not winning in multi-threaded productivity against the i7-6900k and losing practically all gaming benchmarks to the i7-7700k. No surprise there as single-threaded performance is still king in most games. Canuks is no different and Anand omitted gaming benchmarks from its initial review, so its incomplete picture skews perception.

As far as I am concerned, Ryzen delivered pretty much exactly what I was expecting based on leaks and rumors. The most shocking things IMO are the poor memory compatibility at launch and next to no overclocking headroom beyond 4GHz using even high-end conventional cooling.
 
Totally agree and it is really a surprise for me, how adults may write such a bs like a few single persons. Personal attacks and rumors instead of real facts are simply bad style.

 



Well, I expected gaming performance to suffer in the beginning because that happens with every new microarchitecture. When Bulldozer first came out, some games wouldn't even run on it at all. By the time PileDriver came out, my FX-8350 was matching Intel's best and brightest in games like COD4 and ARMA III. What really pisses me off about these reviews is the way the headlines are worded. Sure, the i7-7700K does perform better in games and it is less expensive but that's where its advantages end. In every other metric, it gets beat (sometimes absolutely DEMOLISHED) by Ryzen. In pretty much all situations that the i7-7700K beats the R7-1800X, it also beats the i7-6900K and sometimes even the i7-6950X! That's a bigger knock on Intel than on AMD but writers like this Alcorn character leave that out of the equations. There's also the fact that somehow, all of these "tests" done on this site have the different results than similar, if not identical, tests on other sites. I'll explain in my next post.
 
Anbello262 I have to disagree with you. I own a 144hz 1 ms 1080p monitor. And there is no visibly detectable difference in the fluidity of motion at 60 FPS, 90 FPS, or 144 FPS. They all visibly render the same image. No, the big difference was the quality of the monitor change made the big difference. Just playing 60FPS on a 144HZ monitor was better than a 60HZ monitor. Also, I've been playing FPS for about 20 years, and played in Sol cal league back during the first counter strike days. Also, these higher FPS ranges than 60 FPS are rather dumb considering mainstream gamers use 60HZ anyway, which makes anything over 60 FPS completely useless. if you are going to spend $500 on a CPU to game at 1080... That just sounds stupid and unpractical. Again you are talking about a very small fraction of the population that would want to try to use hardware over 60HZ and 1080p. And again anything over 1080p will render the same give or take 1 or 2 FPS than intel.
 
If I compare the results and conclusions of real engineers (I came from industry and I have since years very close contatcs to bigger server companies and former colleagues) then I see our review not harsh, but as a very real average. The problem is, as mostly, the software and nearly all current engines are Intel optimized. This is logical, because AMD was over years simply not present. Together with a few not perfect optimized things and early-adaptor-mainboards I'm simply not satisfied with this situation.

This isn't perfect for AMD, bad for us reviewers and at the end even worse for all customers, searching for a real and payable alternate. The launch was too fast and nobody got hardware early enough. Not the mainboard companies, cooler factories or others. I see a lot of lost performance in the incomplete and not matured firmware, in the windows compatibility and memory issues. Many things can be solved with patches or better chipset drivers - I hope so.

The comparison of different sites makes no sense, because all boards and CPUs were totally different. My both 1800X are totally different in quality. One sample goes only to 3.9 GHz, the other one to 4.1 GHz. Also the gaming results. Mainboard A runs better than mainboard B in games, but worse in workloads with big datasets. So what you will compare? Believe only sites, used exactly the benchmark suite that was suggested by AMD? Mama, the world is too complex to find ONE global and perfect conclusion/answer. A lot of sites tested with PR0 CPUs, others with older firmware. I got my official hardware only two days before launch, other tested it 2 weeks ago to visit GDC and other events. This was for me one of the most horrible launches ever. No final hardware, bad support and a company in panic mode with really harsh answers on interesting / important questions.
 
This is the same as before. AMD innovates before the industry can catch up. It took a couple of years for bulldozer and its followers to show performance improvements under windows.

The performance vs Intel Broadwell chips shows the muscle, it just ins't optimized for low thread count gaming. Of course most of us won't wait for 6 months+ for software developers to jump on board, and certainly Intel won't be inspiring them to do so. But I think the most compelling chip is one that isn't being talked about much, the R7 1700.

At $329, that's within reach of many gamers' budgets. And not coincidentally that's the same price Intel's been charging for years for it's Core i7 mainstream product.

So AMD gamers are finally part of the club, DDR4, PCI express 3.0, USB 3.1, and a platform upgrade is finally worth it. Is that competitive?
You bet.
 
I'm just waiting for the first AM4 boards with fully functional ECC support to build me a new, alternate workstation. Only just for fun to see, what Ryzen can do in real and how the performance will increase over the next 12 months (or not). It is it worth to do after each major update a new snapshot and write it down after this year as a short review :)
 
I feel I've been noticing the same thing more and more over time. Used to think Toms was a great source of impartial info, but there's such an obvious theme of downplaying AMD, even in the layout of graphs and order in which they present their results that makes one wonder if there isn't some agreement with Intel going on. For me is signals a time to start looking for reviews elsewhere.



 

Let the singer sing his song. If you only like cherry picking and to read, what you want to see, you will be rotating between the sites like a propeller. The truth is not a bitch, only the alternate truth, made by fanboys and their gurus. But for such things we are really the wrong source. 😀
 

The graphs are laid out by benchmark score from fastest to slowest to make it easy to see the different chips' relative standing compared to the nearest similar-performing chips. No bias there, it is just a basic numerical sort with the fastest chips bubbling to the top.

Are you suggesting that you would like results to be presented in randomized order so the best chips for a given benchmark don't systematically hog the top of the graph? That would only make them a pain in the ass to read and far less useful for most people.

When presenting data across several graphs, people expect results to be consistently presented in an order that makes sense. You need to pick one and stick with it.
 
The gaming potential of Ryzen needs to be better assessed with purely compatible titles. The current values obtained are misleading i guess. Curious to see how AMD suggested titles work against 7700K.
 


The suggestions were:
Sniper Elite 4, Battlefield 1, Battlefield Hardline, Star Wars: Battlefront, The Division, Assassin’s Creed: Syndicate, Overwatch, Witcher 3, Dead Rising 4

Try Google to see, who was unsing it finally.... :)
 
Ok, so here are some things that don't add up AT ALL. We have MASSIVE variances in benchmarks between the review on TomsHardware and the reviews elsewhere. Somehow, in every single situation, Toms Hardware's tests either make Intel look better or they make Ryzen look worse. Let's explore this with a couple of examples, and I will only use reputable sites that have been around for a LONG time, not some kid with a YouTube channel. I think that Hard OCP is well respected, also Hardware Canucks and AnandTech should be well-recognised:
r_600x450.png

Ok, so here we are with the Ryzen 1800X which turbos to 4GHz but for some reason, it's only at 3.8GHz here, so what's crippling it? The numbers are also quite inflated compared to what Hard OCP's results are for the same game:
1488169187kcPgB2ioTd_4_5.png

What's also really interesting (and telling) is that their Ryzen CPU is running at 4GHz in the CPU-focused 1280x768 setting of Ashes of the Singularity but only gets an average of 41fps which would seem to be consistent with the minimum frame rate of the "Ryzen 1800X HP" setting so perhaps that is a fair number, but only if Hard OCP is using minimum frame rates. The real problem here is the discrepancy between the results that the two sides have when it comes to the Intel CPUs. How is it that the i7-7700K, OC'd to 5GHz only shows 45.4fps while NON-OVERCLOCKED i7-7700K shows a MINIMUM frame rate of 59fps? Also, how is it that Hard OCP's 6900K OC'd to 4.3GHz only shows 53.1fps while Tom's test with an NON-OVERCLOCKED i7-6900K shows a MINIMUM of 61fps! It's quite apparent that the 1280x768 would be a more CPU-intensive resolution than 1080p especially with the CPU-focused setting. Why didn't Tom's Hardware use this like Hard OCP did? Also, why use the extreme preset? Doesn't that make it more graphics intensive? Some things here don't add up eh? How do overclocked CPUs get lower frame rates in lower resolutions than non-overclocked CPUs when games are most likely to scale well from increased clock speeds? Is it a mystery, or is it something even more sinister than that?

Then of course, there's the wording from the Workstation Section...

"It’s questionable if it’s a good investment if the application in question has been heavily optimized for Intel processors, but there’s no real drawback to its use as a general-purpose CPU."

How is it a questionable investment compared to CPUs that are more than double the cost but never even come close to showing double the performance and are in fact beaten at least once? Regardless of the applications shown, the Ryzen defeated the i7-6900K far more often than not and whenever it was beaten by the i7-7700K, the i7-6900K was also beaten. A very sneaky thing was also done as overclocked Intel CPUs were slipped into the mix to drive the ranking of the Ryzen down on the chart, this is a psychological tactic to give a visual impression of being worse than it is. In AutoCAD 2016 for instance, an i7-6900K overclocked to 3.8GHz was added to the mix with the normal i7-6900K. This was never mentioned in the testing methodology and again, Paul does his passive-aggressive best to make Ryzen look like a disappointment and make the Intel CPUs look like winners:
aHR0cDovL21lZGlhLmJlc3RvZm1pY3JvLmNvbS9YLzEvNjU2OTY1L29yaWdpbmFsLzAxLUF1dG9DQUQtMkQucG5n

"The AutoCAD results remind us of what we reported on the previous page when we looked at output devices. AMD’s Ryzen 7 1800X ends up exactly where you’d think, based on those findings. There’s no real parallelization, but there’s plenty of system memory and cache usage, which turns out to be a combination that puts its performance squarely in the middling category." - In other words, it performs EXACTLY like an Intel CPU that is more than DOUBLE THE COST. Obviously, if the Ryzen loses to the i7-7700K, it is a loss, regardless of whether or not the i7-6900K at more than double the price ALSO loses to the i7-7700K. The overclocked i7-6900K is a clever way here to push the Ryzen down further in ranking while still being mute concerning the non-overclocked i7-6900K.
https://img.purch.com/w/711/aHR0cDovL21lZGlhLmJlc3RvZm1pY3JvLmNvbS9YLzIvNjU2OTY2L29yaWdpbmFsLzAyLUF1dG9DQUQtM0QucG5n
This slide does the same thing, using the overclocked i7-6900K to push the Ryzen from the #3 spot where it could be called one of the leaders into the #4 spot where it justified Paul Alcorn's description of "middling". Again, the overclocked i7-6900K was never mentioned in the test setup, this is either very sloppy or disingenuous. Mind, you, I've never known Tom's Hardware to be sloppy.

In the 3DS Max test, these are the words used:
"As was the case with Blender, the 3ds Max benchmark distinguishes between several different areas, which don’t just include CPU rendering and, consequently, provide a good idea of what the work processes would feel like to the user. AMD’s Ryzen 7 is at a disadvantage when there’s real-time graphics output via the GPU or when only a few cores are used."
- Now, does that sound like an accurate description of a CPU that dominated two of five slides and defeated the i7-6900K in three of five slides? A non-biased way to say sum up the 3DS Max test would have looked more like this:
"As was the case with Blender, the 3ds Max benchmark distinguishes between several different areas, which don’t just include CPU rendering and, consequently, provide a good idea of what the work processes would feel like to the user. AMD’s Ryzen 7 is at a disadvantage when there’s real-time graphics output via the GPU or when only a few cores are used but in situations where many cores are used in a cpu-intensive task, Ryzen dominates." - It's another tactic in writing where the writer is quite willing to condemn a product's shortcomings but is completely unwilling to praise its strengths. The pro-Intel bias is clear.

In the scientific section, when AMD beats Intel, we get a non-dramatic, uninterested assessment like this:
"NAMD is a benchmark for high-performance simulation of large biomolecular systems. All individual tests went Ryzen 7 1800X’s way."

When Intel's VASTLY overpriced i7-6900K and i7-6950X CPUs beat Ryzen (even though Ryzen still beat out the i7-7700K) we get this:

"The Fastest Fourier Transform in the West, or FFTW, is a popular open-source solution to compute one-, two- and three-dimensional DFTs (Discrete Fourier Transforms). The C library makes heavy use of single-precision AVX these days, which proves to be a terrible thing for the Ryzen 7 1800X."

So basically, the word "Terrible" is used when Ryzen gets beat by the stupidly overpriced processors. Interestingly, Ryzen thoroughly defeats Intel's brand-new Kaby Lake architecture in this test that is supposed to be "TERRIBLE" for Ryzen, but not a word is mentioned about it. No, the i7-7700K is only mentioned when it WINS against Ryzen. Again, very telling, could it be that Intel wanted it this way?

So the gist of this review that I get (and Paul, if this wasn't your intention, I hope that you see now that it sure looks that way) is that Paul decided from the beginning that AMD was not going to have a chance here because he was going to take one AMD CPU and put it up against ALL of Intel's CPUs at the same time. So, if the i7-7700K got better frame rates (at gaming resolutions that nobody who pays $500USD for a CPU uses), then Ryzen is a failure. If the R7-1800X loses to the more-than-twice-as-expensive i7-6900K or the more-than-triple-as-expensive i7-6950X, then Ryzen is a failure.

The original thesis of this article is that Ryzen is a failure and then Paul went about proving it. Where is the praise for defeating the Kaby Lake i7-7700K in situations where that happens? This review seems to only acknowledge the Intel CPUs that beat Ryzen, not the Intel CPUs that Ryzen beats. For instance, this review only seems to recognise the i7-7700K's existence in gaming situations where it tends to win. I suppose that Broadwell and Broadwell-E are both phenomenal flops because they're clearly too expensive to even be compared to Ryzen despite the fact that they lose as often as they win against it (if you're looking at reviews on any site other than Tom's Hardware that is).

In this review, Ryzen is made to face every CPU in Intel's stable instead of what should be apples-to-apples. To make things worse, commentary is only made about the Intel CPU that perhaps beats Ryzen because it is completely specialised to that specific test. Again, the Ryzen will not be allowed to win no matter how expensive this review must go to find an Intel chip that will beat it which makes for a horrible comparison. If Ryzen does win, a rather bored commentary is made about it to make the reader skip over it. The way this article has been designed, the reader will think that it's the second coming of Bulldozer. Astonishingly, the price card is quite happily played when the less expensive i7-7700K shows completely trivial leads in gaming. I say trivial because no system builder is going to spend over $300 on a gaming CPU but then get a low-end GPU and game at 1080p, that's just plain stupid.

In this review, the combination of what is said and what is omitted work together to paint a tragically-skewed picture of a phenomenal product. It's really no wonder that people think Intel has some influence on these reviews when one sees reviews like this. I read this review and see a thickly-veiled and very well-put together assassination of what is in fact, the best all-around CPU in the world when the "holy trinity" of performance, price and versatility are taken into account.
There's no way to beat Ryzen in all metrics. In fact, the only way to beat it in most is to have three Intel CPUs costing thousands of dollars and even then, they won't win by much.

Some people are pure gamers and that's fair enough. REAL pure gamers though, game at resolutions that make the CPU completely irrelevant. Some people, like me and I would expect most others on this site, are high-end gamers but we do a lot more with our PC's than just that. I do video editing for instance and my main rig has almost 30TB of drive space (most of it full). For me, having a CPU that can encode video in the background but still have enough power left over for me play a high-end game to pass the time is far more valuable than a few extra FPS especially since I use a 4K display. I'd love to see what kind of gaming numbers the "holy" i7-7700K would get with something like that going on in the background. I sense that it would handle it well, but would take a far harder hit to its fps than Ryzen would.

So there's my "review of the review". I don't speak as an AMD fanboy because I'm not, I'm driven by a hate for the corporate practices of Intel and NVidia, not a love for AMD. I speak as a long-time and well-decorated expert of Tom's Hardware who cannot believe what has happened to this site that was once the best in the world, bar none, for ethical reviews and a wonderful lack of bias. I love this site, I always have and it breaks my heart to see so irresponsible a review like this. I guess Intel's hands reach even into the places that I thought were clean.
 
Your post are tl;dr... Avro...

Is it too much of an impossibility that the NVIDIA TITAN that was used at HardOCP for their gaming test and not the Titan XP that was use for their VR Bench causing that performance difference against Tom's GTX 1080?

And the 3.8Ghz was already explained many times as comparison point between the CPU at the same clock.
 
A lot of text and nothing behind.... To compare results right, the used hardware and clock rates must be 1:1 the same. But explain this iron law some fanboys. Impossible. 😀

I can benchmark only physical existing things and not prices. Prices comes and go, the hardware will be the same. And dude, we tested all CPUs at stock and one time at the same, fixed clock rate that gave AMD more advantages than Intel.

My 2D tests and also AutoCAD shows, that the windows scheduler has problems with Ryzen and that the latency is a bottleneck in some cases. This is (in my opinion) the follow of a not optimized hardware and software, nothing else. Fast launch, too early. As I wrote earlier.
 


Rants like yours makes me think you missed the conclusion;
"It’s hard to recommend the Ryzen 7 1800X over Intel's lower-cost quad-core chips for gaming, especially given the Core i7-7700K's impressive performance. That's not a knock against AMD, specifically. After all, we say the same thing about Intel's own Broadwell-E CPUs. High-end Kaby Lake processors constantly challenge pricier competitors, and the flagship -7700K sells for $350. Even after down-clocking the -7700K to 3.8 GHz, it still beats Ryzen 7 1800X in nearly every game in our suite. Those issues would only be exacerbated on a Ryzen 7 1700X, which operates at lower clock rates.
Conversely, the Ryzen 7 1800X is in its element when you throw professional and scientific workloads at it. It isn't the fastest in every high-end benchmark, but any calculation that factors in value almost assuredly goes AMD's way. For years, Intel has operated with impunity, charging inflated prices for incremental speed-ups. The 1800X’s $500 price tag and competitive performance will no doubt excite power users on a budget. To that end, when we weigh the 1800X’s strong showing in workstation and HPC workloads against its issues with games, we can't help but believe that AMD designed this specific configuration with a datacenter-driven mindset and didn’t optimize it thoroughly for desktops. Much like Intel and Broadwell-E, in fact."
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom