5 ms·

I have the original Threadripper 1950X overclocked to 4.1ghz and I’ve been pretty happy with it. As a developer I can’t find much reason to upgrade 2 years afte

I have the original Threadripper 1950X overclocked to 4.1ghz and I’ve been pretty happy with it. As a developer I can’t find much reason to upgrade 2 years after its release. Single core improvements (from AMD or Intel) aren’t game-changing. More cores won’t do anything for me (.NET Core, angular, etc) I’ve considered switching back to intel for their 5ghz processors but based on benchmarks, I wouldn’t see anything but marginal improvements. All in all, competition is good, but I wish they’d focus on single core performance improvements.

ses1984 7y ago
I can still put more cores to work, compiling stuff. I have a 6 core quad channel intel and I've been waiting for this threadripper to come out.
- rbanffy 7y ago
  Maybe, but how often do people compile their software?
  - ses1984 7y ago
    If you're editing it, a lot.
astrodust 7y ago
If you ever get into Docker/Kubernetes, being able to spin up a local cluster for testing is a pretty big deal. Having that running in the background and not impacting your usual workload is huge. I've been testing a Kubernetes cluster built on 32-core AMD servers and it's unreal the workloads you can throw at them. I'm used to 4 or 8-core Xeon chips and this is a whole different game.
- marmaduke 7y ago
  How much of that is core count though? I've used minikube several times and it didn't seem to matter much how many cores it had (beyond 4)
  - astrodust 7y ago
    For simple things it doesn't matter, but if you're bringing up a complex application that has a dozen server components and putting it under load, it does matter.
- rbanffy 7y ago
  That's true, but memory/IO is a much bigger issue in this scenario than processing power. While running a cluster in VMs, I'm frequently limited by IO (and if your VMs use more memory than it's physically available, you'll quickly descent into swap hell) well before my CPUs register a significant uptick in usage. And, mind you, my workstation is 4 year old Xeon with 64 gigs of memory and reasonably fast (but not amazing) SATA SSD. On a side note, I work more and more from a couple of i3 and i5 laptops and only use the workstation to do heavy lifting tasks, such as replicating these more exotic setups.
  - spamizbad 7y ago
    I really hope the 3000-series Threadripper is 8-channel like its server counterpart. 8 16GB DIMMs gets you 128GB, and a significant amount of memory bandwidth. You also have enough PCI-E lanes to either stripe or partition your I/O in such a way to work around any major bottlenecks.
    - AstralStorm 7y ago
      There will be a new socket for 8 channel ones. sWRX8. These are not unlocked and essentially baby version of Epyc.
    - kllrnohj 7y ago
      It's definitely going to start at quad-channel + 64 PCI-E lanes like the existing sRT4 platform as it's almost certainly going to have sTR4 socket-compatible SKUs. Rumors are there will also be an 8-channel + 128 lane one as well. Supposedly the quad-channel will be the platform with overclocking support, the more "enthusiast" platform, and the 8-channel one will really more be the workstation platform. https://www.gamersnexus.net/news-pc/3510-hw-news-threadripper-8-channel-4-channel-leaks https://www.gamersnexus.net/news-pc/3510-hw-news-threadrippe...
    - bluecalm 7y ago
      My biggest hope is that you will be able to get 256 or 512GB of RAM with those. The biggest limitation for the previous one for me in comparison to Xeons was that you could only got 128GB of RAM.
arcticbull 7y ago
They're doing both. AMD has ramped single-core performance, at least from an IPC perspective, massively. From Excavator (2015) to Ryzen we saw a 52% increase in IPC. From Ryzen 1000 to 2000 (Zen+) we saw a 3% increase in IPC. From Ryzen 2000 to 3000 (Zen2) we saw a 15% increase in IPC. From 2015 to 2019 we saw a total IPC boost on the AMD side of over 80%, and an increase in max core count in the desktop line of 433%, from 6 to 32. I'm okay with this progress :)
- GiorgioG 7y ago
  The progress is ok, just incremental. AMD had been playing catch up, until Ryzen/Threadripper. Just for comparison, here are the numbers from my GeekBench 4 run: Single-Threaded: 4,746 Multi-Threaded: 34,586 According to a leaked benchmark, the Threadripper 3000's numbers are: Single-Threaded: 5,519 Multi-Threaded: 68,279 The multithreaded benchmark is 2x, that's a no-brainer since it's likely to have 32 cores vs the 1950x's 16 cores. Now, I will say that TR3000 benchmark is not overclocked (3.6ghz.) But from what I've read, it seems like there's not much room for these latest chips to be overclocked. So, despite being the 3rd iteration, the TR3000 is only 16% faster in single-threaded benchmarks than my 1950X. Intel's 9900K (overclocked) gets a single-threaded GB4 score of ~7,000. That (or its successor) may be my next machine.
  - x3sphere 7y ago
    Honestly, that seems rather low. It looks like a lot of Ryzen 3900X's are doing well over 6000 on Geekbench? Ie. https://browser.geekbench.com/v4/cpu/14151649 https://browser.geekbench.com/v4/cpu/14151649 Is the Threadripper 3000 test also from Geekbench V4? I noticed they added V5. I don't see why it should be so far behind the Ryzen in single thread... unless the boost isn't working properly or something, could be disabled if it's a test chip.
    - GiorgioG 7y ago
      Yep, it's from GBv4. It could be that the 3900X is clocked higher than the TR3000. Given that they're packing more cores on the chip, it's possible that they can't dissipate heat as well and limit the single-core top-end speed on the TR3000s more than on the Ryzen 3000 series.
agumonkey 7y ago
I'd love to have even a 1st gen ryzen to know what it's like to be able to compile massively without feeling it. Right now my poor x201 overheats to death if I use rustc carelessly.
kllrnohj 7y ago
The biggest upgrade over Threadripper 1950X is that the new ones are no longer NUMA. So performance across the cores will be much more consistent even if the peaks are not that much bigger. If your development involves compiling code then the 2x larger L3 is also going to result in huge improvements to code compilation speed as seen in the Epyc Rome reviews. Example: https://www.phoronix.com/scan.php?page=article&item=amd-epyc-7502-7742&num=5 https://www.phoronix.com/scan.php?page=article&item=amd-epyc... - compare the 7601 vs. the 7502. Both are 32c/64t parts. Both have close-enough base clocks & turbo clocks. But the new one (7502) absolutely smokes the old one (7601) at both kernel compile in GCC and LLVM compile. Compilers love that L3 it seems. All signs point to Threadripper 3000 still being sTR4 compatible, so that means you could drop in a CPU that gives you +15% single core performance vs. previous gen's best case, no more NUMA domains bringing more consistent performance, lower power usage, and double the L3 for much faster compiles. That's pretty awesome particularly given we're only talking a ~2 year gap between the products.
- loeg 7y ago
  Wait in what sense are the new Threadrippers no longer NUMA? Edit: Found the Zen 2 Epyc marketing materials that describe this. Yes, apparently memory access on a single socket is uniform, in that all memory access is indirected over the IO chiplet[1]! This may hurt best-case access latency for NUMA-aware workloads? Just speculating. It's not like non-local caches go away with uniform DRAM access latencies — unless L1/L2/L3 are also non-local to the core and indirected behind the IO chiplet. Which would be really surprising. [1]: https://www.servethehome.com/amd-epyc-7002-series-rome-delivers-a-knockout/4/ https://www.servethehome.com/amd-epyc-7002-series-rome-deliv...
  - kllrnohj 7y ago
    In the definition sense? It's a single memory controller on the IO die & one memory domain. The chiplets don't have their own memory controllers. By contrast Threadipper 1 & 2 had multiple memory controllers. It was 2x dual-channel controllers. As such they were full on real NUMA, just like a multi-processor system.

ses1984 7y ago

I can still put more cores to work, compiling stuff. I have a 6 core quad channel intel and I've been waiting for this threadripper to come out.

rbanffy 7y ago
Maybe, but how often do people compile their software?
- ses1984 7y ago
  If you're editing it, a lot.

astrodust 7y ago

If you ever get into Docker/Kubernetes, being able to spin up a local cluster for testing is a pretty big deal. Having that running in the background and not impacting your usual workload is huge. I've been testing a Kubernetes cluster built on 32-core AMD servers and it's unreal the workloads you can throw at them. I'm used to 4 or 8-core Xeon chips and this is a whole different game.

marmaduke 7y ago
How much of that is core count though? I've used minikube several times and it didn't seem to matter much how many cores it had (beyond 4)
- astrodust 7y ago
  For simple things it doesn't matter, but if you're bringing up a complex application that has a dozen server components and putting it under load, it does matter.
rbanffy 7y ago
That's true, but memory/IO is a much bigger issue in this scenario than processing power. While running a cluster in VMs, I'm frequently limited by IO (and if your VMs use more memory than it's physically available, you'll quickly descent into swap hell) well before my CPUs register a significant uptick in usage. And, mind you, my workstation is 4 year old Xeon with 64 gigs of memory and reasonably fast (but not amazing) SATA SSD. On a side note, I work more and more from a couple of i3 and i5 laptops and only use the workstation to do heavy lifting tasks, such as replicating these more exotic setups.
- spamizbad 7y ago
  I really hope the 3000-series Threadripper is 8-channel like its server counterpart. 8 16GB DIMMs gets you 128GB, and a significant amount of memory bandwidth. You also have enough PCI-E lanes to either stripe or partition your I/O in such a way to work around any major bottlenecks.
  - AstralStorm 7y ago
    There will be a new socket for 8 channel ones. sWRX8. These are not unlocked and essentially baby version of Epyc.
  - kllrnohj 7y ago
    It's definitely going to start at quad-channel + 64 PCI-E lanes like the existing sRT4 platform as it's almost certainly going to have sTR4 socket-compatible SKUs. Rumors are there will also be an 8-channel + 128 lane one as well. Supposedly the quad-channel will be the platform with overclocking support, the more "enthusiast" platform, and the 8-channel one will really more be the workstation platform. https://www.gamersnexus.net/news-pc/3510-hw-news-threadripper-8-channel-4-channel-leaks https://www.gamersnexus.net/news-pc/3510-hw-news-threadrippe...
  - bluecalm 7y ago
    My biggest hope is that you will be able to get 256 or 512GB of RAM with those. The biggest limitation for the previous one for me in comparison to Xeons was that you could only got 128GB of RAM.

arcticbull 7y ago

They're doing both. AMD has ramped single-core performance, at least from an IPC perspective, massively. From Excavator (2015) to Ryzen we saw a 52% increase in IPC. From Ryzen 1000 to 2000 (Zen+) we saw a 3% increase in IPC. From Ryzen 2000 to 3000 (Zen2) we saw a 15% increase in IPC. From 2015 to 2019 we saw a total IPC boost on the AMD side of over 80%, and an increase in max core count in the desktop line of 433%, from 6 to 32. I'm okay with this progress :)

GiorgioG 7y ago
The progress is ok, just incremental. AMD had been playing catch up, until Ryzen/Threadripper. Just for comparison, here are the numbers from my GeekBench 4 run: Single-Threaded: 4,746 Multi-Threaded: 34,586 According to a leaked benchmark, the Threadripper 3000's numbers are: Single-Threaded: 5,519 Multi-Threaded: 68,279 The multithreaded benchmark is 2x, that's a no-brainer since it's likely to have 32 cores vs the 1950x's 16 cores. Now, I will say that TR3000 benchmark is not overclocked (3.6ghz.) But from what I've read, it seems like there's not much room for these latest chips to be overclocked. So, despite being the 3rd iteration, the TR3000 is only 16% faster in single-threaded benchmarks than my 1950X. Intel's 9900K (overclocked) gets a single-threaded GB4 score of ~7,000. That (or its successor) may be my next machine.
- x3sphere 7y ago
  Honestly, that seems rather low. It looks like a lot of Ryzen 3900X's are doing well over 6000 on Geekbench? Ie. https://browser.geekbench.com/v4/cpu/14151649 https://browser.geekbench.com/v4/cpu/14151649 Is the Threadripper 3000 test also from Geekbench V4? I noticed they added V5. I don't see why it should be so far behind the Ryzen in single thread... unless the boost isn't working properly or something, could be disabled if it's a test chip.
  - GiorgioG 7y ago
    Yep, it's from GBv4. It could be that the 3900X is clocked higher than the TR3000. Given that they're packing more cores on the chip, it's possible that they can't dissipate heat as well and limit the single-core top-end speed on the TR3000s more than on the Ryzen 3000 series.

agumonkey 7y ago

I'd love to have even a 1st gen ryzen to know what it's like to be able to compile massively without feeling it. Right now my poor x201 overheats to death if I use rustc carelessly.

kllrnohj 7y ago

The biggest upgrade over Threadripper 1950X is that the new ones are no longer NUMA. So performance across the cores will be much more consistent even if the peaks are not that much bigger. If your development involves compiling code then the 2x larger L3 is also going to result in huge improvements to code compilation speed as seen in the Epyc Rome reviews. Example: https://www.phoronix.com/scan.php?page=article&item=amd-epyc-7502-7742&num=5 https://www.phoronix.com/scan.php?page=article&item=amd-epyc... - compare the 7601 vs. the 7502. Both are 32c/64t parts. Both have close-enough base clocks & turbo clocks. But the new one (7502) absolutely smokes the old one (7601) at both kernel compile in GCC and LLVM compile. Compilers love that L3 it seems. All signs point to Threadripper 3000 still being sTR4 compatible, so that means you could drop in a CPU that gives you +15% single core performance vs. previous gen's best case, no more NUMA domains bringing more consistent performance, lower power usage, and double the L3 for much faster compiles. That's pretty awesome particularly given we're only talking a ~2 year gap between the products.

loeg 7y ago
Wait in what sense are the new Threadrippers no longer NUMA? Edit: Found the Zen 2 Epyc marketing materials that describe this. Yes, apparently memory access on a single socket is uniform, in that all memory access is indirected over the IO chiplet[1]! This may hurt best-case access latency for NUMA-aware workloads? Just speculating. It's not like non-local caches go away with uniform DRAM access latencies — unless L1/L2/L3 are also non-local to the core and indirected behind the IO chiplet. Which would be really surprising. [1]: https://www.servethehome.com/amd-epyc-7002-series-rome-delivers-a-knockout/4/ https://www.servethehome.com/amd-epyc-7002-series-rome-deliv...
- kllrnohj 7y ago
  In the definition sense? It's a single memory controller on the IO die & one memory domain. The chiplets don't have their own memory controllers. By contrast Threadipper 1 & 2 had multiple memory controllers. It was 2x dual-channel controllers. As such they were full on real NUMA, just like a multi-processor system.