AMD Ryzen Threadripper 1900X CPU Review

Overclocking, Infinity Fabric & Test Setup

The Architectural Bits

The Ryzen 7 series features a single-die design, whereas AMD's Threadripper models employ a quad-die package with just two of its dies active. That means the 8C/16T 1900X effectively wields a dual-die arrangement with four active cores in each die.

Our first slide below shows the alignment of AMD's 16C/32T Ryzen Threadripper 1950X. Each die contains a pair of four-core CPU complexes (CCXes) that incur increased latency when they communicate with the neighboring CCX (denoted as CCX0 and CCX1). Another layer of latency comes into play when they communicate with the CCXes resident on the second die (marked as Die1). Simplified, the greater distance between dies means that die-to-die latency is much higher than the latency between two CCXes resident on the same piece of silicon.

The larger Threadripper models distribute active cores across both CCXes inside each die. The second slide shows how AMD disables one core per CCX (blocked out in blue) to create the smaller 12C/24T 1920X model.

AMD takes an entirely different tack with its 1900X, as seen on the third slide. In a bid to eliminate one layer of latency, AMD confines the 1900X's active cores to a single CCX inside each die. The adjustment makes sense; spreading the cores evenly across all four CCXes increases the chance of incurring latent communication with neighboring CCXes. 

Disabling two entire CCXes also has other implications, though. Inactive cores, provided they are near active cores, can absorb excess heat, potentially improving overclockability. Case in point: it's common to achieve lower overclocks on the 16C 1950X than the 12C 1920X. Consequently, the 1900X's clustered cores should reduce latency, eliminating the CCX-to-CCX delay entirely and only leaving us with die-to-die latency. But they could also potentially hamper overclocking.

The Infinity Fabric Breakdown

A few quick tests with SiSoftware's Sandra Multi-Core Efficiency test illustrate the consequence of AMD's design decisions. Flipping the 1900X into Creator Mode, with all eight cores active, results in three distinct layers of latency. In contrast, the 1950X in Creator Mode has four layers.

Switching the 1900X into Game Mode disables one entire die, leaving us with a 4C/8T processor that has only two layers of latency. Again, the 1950X in Game Mode has a third layer that affects performance. But the 1900X in Game Mode also achieves the lowest fabric bandwidth in our line-up of AMD models.

Finally, the two disabled dies remove a total of 16MB of L3 cache from the 1900X. That means it offers half of the multi-threaded cache bandwidth of the 1950X in Game and Creator Mode. Incidentally, the 1900X demonstrates less multi-threaded throughput than Ryzen 7 1800X in Game Mode. But the 1800X also proffers eight cores with simultaneous multi-threading, while the 1900X in Game Mode drops to a 4C/8T processor with only 8MB of L3 cache. Many games are sensitive to memory and cache performance, so it will be interesting to see how that plays out in our game testing.

Overclocking

Overclocking the 1900X was an exercise in simplicity. We merely adjusted the data rate to DDR4-3200 and set timings at 14-14-14-34. We increased Vcore to 1.39V, well below AMD's recommended maximum of 1.45V, and adjusted the SoC voltage to 1.1V. This proved stable up to 4 GHz during extended stress tests. However, even in the face of unsafe voltages, we were unable to attain a stable 4.1 GHz overclock to match our efforts with the 12-core Threadripper 1920X. Dialing back the memory frequency didn't help, either. Considering that AMD supposedly selects the top 5% of its dies for Threadripper CPUs, you might assume that the clustered active core arrangement comes into play. We only see a 100 MHz reduction, so it's more likely that our retail sample is simply on the lower end of the bell curve.

It's notable that a 4 GHz overclock might actually hamper the 1900X's performance in lightly threaded workloads, since we lose the benefit of a quad-core 4.2 GHz XFR boost. 

Test Systems

Test System & Configuration
Hardware
AMD Socket SP3 (TR4)
AMD Ryzen Threadripper 1950X, 1920X, 1900X
Asus X399 ROG Zenith Extreme
4x 8GB G.Skill Ripjaws V DDR4-3200 @ 2666 and 3200 MT/s

Intel LGA 2066
Intel Core i7-7820X
MSI X299 Gaming Pro Carbon AC
4x 8GB G.Skill Ripjaws V DDR4-3200 @ 2666 and 3200 MT/s

AMD Socket AM4
AMD Ryzen 7 1800X
MSI X370 Xpower Gaming Titanium
2x 8GB G.Skill RipJaws V DDR4-3200 @ 3200 MT/s

Intel LGA 1151
Intel Core i7-7700K
MSI Z270 Gaming M7
2x 8GB G.Skill RipJaws V DDR4-3200 @ 2666 and 3200 MT/s

All

EVGA GeForce GTX 1080 FE
1TB Samsung PM863
SilverStone ST1500, 1500W
Windows 10 Creators Update Version 1703
Corsair H115i

MORE: Best CPUs

MORE: Intel & AMD Processor Hierarchy

MORE: All CPUs Content

This thread is closed for comments
28 comments
    Your comment
  • rcrossw
    My 1900x is on an Asus Prime X399-A and running stable at 4190 Ghz. Temp 44 C. Prior to the 403 Bios it ran at 4225 Ghz.

    This is to let others know what I was able to do with the 1900x. I use a Ryzen 1700 for gaming. I do Photo work and Ballistics, on the 1900x. For what I use it for it is superb product. One last thing - I run both systems at 4K Res. Better on my older eyes.
  • TechyInAZ
    Excellent review. I've been waiting for this review for a while since it's the strangest threadripper CPU in the family. Specifically i was wondering if they were going to do two cores per CCX, glad they decided to do just all 4 in one CCX for better latency.
  • Kawi6rr
    Is Threadripper a gaming CPU? If not then why do you test in so many games? Doesn't make sense.
  • PaulAlcorn
    1864409 said:
    Is Threadripper a gaming CPU? If not then why do you test in so many games? Doesn't make sense.


    31 tests for applications. Some tested with both native CPU processing and OpenCL acceleration. Also, four synthetics that measure key performance traits.


    10 games, four synthetics.
  • antonysg77
    For a normal user who also plays games, Ryzen 7 1700 is a powerful processor, which is power efficient and also comes with its own cooling solution. Also, both the processor and motherboard are very reasonably priced.
  • mapesdhs
    1920539 said:
    31 tests for applications. Some tested with both native CPU processing and OpenCL acceleration. Also, four synthetics that measure key performance traits that are applicable to some apps.


    Bit odd basing the conclusion on tests that are not by the initial introduction representative of the target market for the product. Why does everything have to be about gaming?? If the CPU isn't aimed at gamers in the conventional sense then surely it makes more sense to test it based on the kind of task it is aimed at? For example, setup an X399/1900X system with four GPUs for CUDA in AE or somesuch, compare it to the same GPU config on an X299 board, how do they behave? Efficiency, power consumption, render times, stability during an intense render, etc.

    At the very least do some tests at 4K while streaming and show how the systems compare under such a scenario, such as GN has done for various CPU comparisons.

    Ian.
  • cryoburner
    1864409 said:
    Is Threadripper a gaming CPU? If not then why do you test in so many games? Doesn't make sense.


    It makes sense in that they can show that to people who might be considering going with it for a gaming system. Some people tend to think that just because some piece of hardware is more expensive that it will be better for gaming, when in reality that hardware may cost more because it adds features that don't even provide much benefit to games. The 1900X enables quad channel memory with a higher maximum memory limit, but games won't benefit from that, and 16GB of dual channel memory should work just as well for years to come. Likewise, a gaming system won't likely see much benefit from having a CPU with 64 PCIe lanes. Someone wanting an 8 core processor for a gaming system would likely get comparable performance by overclocking a Ryzen 1700 on an X370 motherboard for several hundred dollars less. The same goes for other HEDT processors with lots of cores like the other Threadripper parts and Intel's equivalents. For gaming, those extra cores won't likely provide any benefit, and will likely only make it harder to keep the chip cool,resulting in lower clock rates if anything. Of course, there will also be some people who want those extra hardware features for specific tasks other than gaming, but may want to be able to game on the system as well.
  • anbello262
    1864409 said:
    Is Threadripper a gaming CPU? If not then why do you test in so many games? Doesn't make sense.


    This site is aimed mainly at gamers, so it makes sense to base the conclusion on gaming performance. You just seem to be upset that "someone might read this the wrong way and think AMD sucks", as if our job would be caring for the companies best interests.
    They can do their own marketing, the job of Tom's is to give us all the information in the most scientifically accurate way possible, and then sum it up with a conclusion aimed at their readers, who are mostly gamers.
    If you are not a gamer, then you can just read the pages with the productivity apps, and just ignore the conclusion (since it will not apply to you). There is no misrepresentation or false information anywhere in the article
  • PaulAlcorn
    Well, the conclusion also takes application performance into account.

    Quote:
    "The real competition happens in our application workloads."


    Also, there are seven application price efficiency charts in the conclusion.
  • Nintendork
    BILLY GATES
    Dunno if you forgot than with any TR chip you will get the full 64pcie lanes + ECC support. Where can you find that even on the 2K i9? Well, you won't.
  • rhysiam
    1594419 said:
    7820x is better in every way. AMD needs to stay in their lane and that is the lowend-midrange builds.

    Unless you need more than 28 PCIe lanes or ECC RAM. In which case Intel locks you out unless you pay much higher prices.

    You've 'missed the point of this CPU entirely! If you care primarily about value for-money CPU performance this is a terrible CPU, in fact, one of the worst on the market. If, on the other hand, your primary concern is large amounts of ECC ram and/or many PCIe lanes for the lowest possible price, this will be for many the best purchase on the market. It's a niche product for sure, but for the few who want it, this is a fantastic product.
  • DerekA_C
    i see 8k gaming a thing or 4k 144hz possible with that many lanes possibly with vega refresh or 4 1080ti.
  • Gillerer
    I think that for CPU reviews' game benchmarks, it would be interesting to have a graph indicating how many cores are at full or near full utilization - indicating CPU bound scenario, or risk of getting CPU bound at certain points during gaming.
  • mdd1963
    You'd almost have to be retarded to intentionally choose a 1900X over an R7-1700...' seriously. :)
  • Gillerer
    46152 said:
    You'd almost have to be retarded to intentionally choose a 1900X over an R7-1700...' seriously. :)


    "almost have to be retarded", with the stress on the word "almost".

    Even if the performance of the 1900X is lacking, there are advantages to the X390 platform: more memory, more PCIe, more NVMe, NVMe RAID, 10Gb NIC available built-in to some motherboards. If you need or want those, but don't need extra cores, the 1900X makes sense.
  • Wisecracker
    54046 said:
    How much did Intel pay for this review??
    :lol:
    You have a point, though _:D_ This is not Ryzen TR 1900X Vs. Ryzen1800x ... or i7-7820X ... it's really AMD Socket sTR4 Vs Chipzilla('s X Next-99 or X Same-As-The-Last-99 Chipset)

    A TR 1900X/X399 Combo is $840US and in no surprise, the i7-7820X/X299 Combo is $5 less. oh, darn. THG whiffed.

    It's Intel X Old-And-Busted-99 Vs AMD's X Eat-Our-Dust-99 Motherboards. They're the same, except where they are not, and except where they're going. The irony that should not be missed is AMD is using Intel's own playbook against them (want more jam? ... upgrade to a 1950X or ? ) plus winning price points, with a long history of strong support for their high-end platforms.

    AMD is now capable of 'cluster-ing' and 'glue-ing' (HA!) on 4,096-landing substrate whatever suits their fancy, and has partnered in the engineering/design of their 'platform/controller hub' and X399 motherboard with AsMedia. Works for me. Show me the money. 4-6-8 CCX ( 8-12-16 ?) or 120 PCIe lanes works for me, too. I'd prefer a SIMD Engine Array with the compute power of an R9 280X to 32 cores / 64 threads but I'm not paid to make that call ...

    My best next-up guess for sTR4 is 2xRR, or 8/16 with 20 CUs (or nCUs, Lexa GCNnCUs? whatever :pt1cable: ) at $370US --- with the same OpenCL compute as the i7-7820X with a GTX 1080.

    That is, if the i7-7820X does OpenCL compute by then
    ('PCtel Mark' need not apply :ouch: )
  • LeonVolcove
    I did not see the power consumption page?
  • SBSExtensions
    Too bad there were no workstation (i.e. SolidWorks CPU & GPU) benchmarks. Could you use the 1900X in future benchmarks against other processors? Thanks.
  • jabliese
    You should add a tag line to every AMD review, "And, of course, the AMD motherboard is lacking Intel's ME." That's a decent selling point.
  • mapesdhs
    1426528 said:
    This site is aimed mainly at gamers, so it makes sense to base the conclusion on gaming performance. ...


    Even though the article opens by stating the product isn't aimed at gamers. Without the context of the conclusion itself being solely aimed at gamers, comments such as, "But 1900X just isn't much more compelling than Ryzen 7 1800X, ...", make no sense at all. Ditto, "... so professionals on the hunt for overall performance may favor Intel's Skylake-X chip.", when there were no relevant use cases even tested. Also, "Thus, we aren't particularly attracted to Threadripper 1900X."; who's the "we" here? Gamers? The conclusion section doesn't state this, it reads like an overall conclusion. I note my earlier post had several down votes; anyone care to respond with actual arguments as opposed to an easy page click? There are some very sensible use cases for this CPU, but the article reads like it's a pointless product because it's so focused on gaming, while talking about a product the article intro says isn't aimed at gamers. No 4K, no multi-GPU, no pro tasks that can exploit what the product can really to those for whom it's a very good fit indeed.

    Ian.
  • lotusmotors
    205977 said:
    BILLY GATES Dunno if you forgot than with any TR chip you will get the full 64pcie lanes + ECC support. Where can you find that even on the 2K i9? Well, you won't.

    Not a single Threadripper owner even uses ECC RAM because it's too slow. ECC is pointless for client PC anyways and only people bringing it up are non-owners.
  • mapesdhs
    2578136 said:
    Not a single Threadripper owner even uses ECC RAM because it's too slow. ECC is pointless for client PC anyways and only people bringing it up are non-owners.


    Spoken by someone who presumably doesn't do a task that needs ECC and hence isn't aware of its relevance. CUDA accelerated financial transaction processing, for example. 1900X with four GPUs and ECC would be ideal for that; ECC is more important than uber speed to those who need it. You claim it's "too slow"; please supply some data *on this platform* which supports this assertion for relevant tasks.

    Ian.
  • wiyosaya
    117741 said:
    1426528 said:
    This site is aimed mainly at gamers, so it makes sense to base the conclusion on gaming performance. ...
    Even though the article opens by stating the product isn't aimed at gamers. Without the context of the conclusion itself being solely aimed at gamers, comments such as, "But 1900X just isn't much more compelling than Ryzen 7 1800X, ...", make no sense at all. Ditto, "... so professionals on the hunt for overall performance may favor Intel's Skylake-X chip.", when there were no relevant use cases even tested. Also, "Thus, we aren't particularly attracted to Threadripper 1900X."; who's the "we" here? Gamers? The conclusion section doesn't state this, it reads like an overall conclusion. I note my earlier post had several down votes; anyone care to respond with actual arguments as opposed to an easy page click? There are some very sensible use cases for this CPU, but the article reads like it's a pointless product because it's so focused on gaming, while talking about a product the article intro says isn't aimed at gamers. No 4K, no multi-GPU, no pro tasks that can exploit what the product can really to those for whom it's a very good fit indeed. Ian.

    I have to agree with your comments. Someone ought to come up with a Finite Element benchmark as the FE type workload is something that benefits from a wide memory bus.
  • Wisecracker
    117741 said:
    ... comments such as, "But 1900X just isn't much more compelling than Ryzen 7 1800X, ...", make no sense at all. Ditto, "... so professionals on the hunt for overall performance may favor Intel's Skylake-X chip.", when there were no relevant use cases even tested. ...


    Pretty much this. But, Anandtech to the rescue (kinda) ..

    Though - it might be for a good reason. Ryzen DT gives Chipzilla HEDT (KL-X, SL-X, BW-E, HW-E) a good run for the money, and AMD X-399 HEDT blows-up Intel's X-299 price structure and HEDT going forward.