If you've never heard of the Ryzen 5 3500X yet, it is likely due to the fact that, depending on the location, this CPU is mainly sold and marketed to OEMs and system integrators. It was originally reported as a purely OEM part for the Chinese market, but then also in other regions such as India. It turns out that you can buy the 3500X practically anywhere, although conditions vary from region to region.
In China, it appears to be a regular retail product that comes in a box with a cooler and everything. In Australia, you can only purchase the 3500X on a pre-built PC. What still raises the question of how well it works and what has changed compared to the R5 3600?
We heard over channels that there are some frames you have to jump through to purchase AMD's 3500X as a system integrator. You need to tell them your plans, the type of system you want to build, and the volume you're likely to move. We also heard that they are only sold in large quantities, i.e. over 1,000 chips.
However, the Ryzen 5 3500X has been available on AliExpress for over a month. The second they were listed, we bought one for $ 157. It's been sitting on our desk for a couple of weeks now, so we thought we'd better test this thing or it just wouldn't happen.
Let’s quickly review the 3500X’s specifications …
Like the 3600, it contains half a dozen Zen 2 cores that are clocked at 3.6 GHz for the base. Depending on the workload, they can instantly clock up to 4.1 GHz, just 100 MHz lower than the 3600. The main difference is the lack of simultaneous multithreading (SMT) support for the 3500X, where the R5 3600 contains 6 packs -cores / 12 threads, the 3500X only has 6 threads, similar to the Core i5-9400F.
Is it worth canceling SMT support for a ~ 30% discount off the R5 3600? (Even in Australia, this depends on your location.) For buyers based in the United States, this is only a 20% discount. We doubt that the savings will be worth it, in any case we will know shortly. For testing, we have the usual set of game and productivity benchmarks.
All CPUs were tested with 16 GB DDR4-3200 CL14 memory and an RTX 2080 Ti to reduce possible GPU bottlenecks. All Windows 10 patches, games, drivers, and BIOS updates available at the time of testing were installed. Let's get to the results.
First, we have the Cinebench R20 multi-core results and here the 3500X doesn't look particularly good. Yes, it's still faster than the Core i5-9400F, but slower than even the first generation Ryzen 5 1600X. It's 6% slower than the R5 2600 and a whopping 27% slower than the R5 3600.
Of course, the 3500X may not be the best solution for those who want to run core-heavy applications.
The advantage of the Ryzen 5 3500X over the similarly expensive Ryzen parts of the first and second generation (1600 and 2600 are 6-core / 12-thread processors) is the improved single-core performance of these Zen 2 cores.
In single-core Cinebench, it essentially corresponds to the Core i7-8700K and is therefore 25% faster than the R5 2600 and even 9% faster than the Core i5-9400F.
The SMT implementation of AMD for Ryzen first and second generation was not particularly useful for compression. Therefore, the 3500X can keep up with the processors of the series 1600 and 2600. That said, it's 22% slower than the R5 3600, which means a relatively large drop in performance.
AMD's SMT has always worked very well on decompression work, so the 3500X is blown away as it can only compete with the Intel Core i5 processors in this test. This time it was 36% slower than the 3600 and 25% slower than the 2600.
As we saw when testing with Cinebench, the 3500X is not the one for those looking for a budget productivity processor. In the V-Ray benchmark we see that the 3500X was 6% slower than the old Ryzen 5 1600 and almost 20% slower than the R5 2600.
When you run the Corona benchmark, it looks worse, much worse. Here the 3500X can only keep up with the Core i5-9400F, which means that it is 27% slower than the R5 1600 and 43% slower than the R5 2600. Compared to the R5 3600, we're talking about a 60% increase in rendering time, so creators on a budget are much better off with a Ryzen 2600 or similar.
We find a similar story when testing with Blender. The 3500X took 14% longer to do the workload than a first-generation Ryzen 5 1600, which means that it was much slower than the R5 2600 and R5 3600.
Removing SMT support will significantly decrease the 3500X's energy efficiency. For example, if it was 14% slower than the first generation R5 1600, the overall system utilization is only 10% lower. This number includes all system components, but you would still expect the newer 7 nm part to do much better. Here, too, power consumption hardly plays a role, since the total system load of 141 watts is nothing for a core-heavy workload.
Time for some game benchmarks and we start with Assassin's Creed: Odyssey at 1080p using the very high quality settings. As a reminder, we use a high-end RTX 2080 Ti for testing.
Here the 3500X with the Core i5-9400F is comparable with the Ryzen 5 parts of the first and second generation. That said, performance is good, but we feel that it won't be long before 6-thread processors like the 3500X and 9400F take the Core i5-7600K's path.
ACO is a bit more GPU-bound at 1440p and this helps the 3500X as it can't keep up with the Ryzen 5 2600X and really delivers similar performance to the R5 3600. The performance from the middle to the top end is pretty similar under these more GPU limited test conditions.
Battlefield V is an example where 6 or even 8 threads are no longer enough. Here the 3500X delivered a level of performance similar to that of a Core i7-7700K and a Core i5-9400F, and although the average frame rate looks pretty good, the 1% low performance suffers.
Granted, the game is still very playable and much better than the quad-core 7600K, but you'll see significantly larger frame dips compared to the 12-threaded and better CPUs.
The 1% low performance was improved by 11% with the R5 1600 and by 14% with the R5 2600. However, it is the R5 3600 that really increases things and increases the 1% low power by almost 30%.
The situation does not change at 1440p, but it only gets worse for the 6-core processor. Here the 3500X still suffers from a sharp 1% drop in performance, and as a result the R5 1600 was 25% faster, which is not a good result for the affordable 3rd generation Ryzen processor.
Shadow of the Tomb Raider is not quite as CPU demanding as the first two titles, which is why the 3500X performs quite well. It can only keep up with the 2600X and 9400F, but overall it's a pretty good result.
The margins are similar at 1440p, the 3500X slips slightly behind the 2600X and 9400F, but again the overall performance was decent.
The Ryzen 5 3500X also has some problems in Division 2 and has a poor performance of 1% compared to the first generation Ryzen 5 1600. Again, we see that the average frame rate performance is quite good, but the 1% low performance that you should focus on here.
Even at 1440p, the 3500X falls behind the R5 1600, if only by a few frames, but you wouldn't expect a 6-core Zen 2 processor to lag behind part of the first generation.
Far Cry New Dawn is the least CPU-demanding game we've tested so far. Therefore, it is not surprising that the 3500X does not have a poor 1% performance here. In this regard, it's actually a little better than the 3600. The overall performance was the same and not much better than the 2nd generation Ryzen.
At 1440p we see the same behavior, so for older and less demanding titles the 3500X with six threads is perfectly fine, no surprises.
Hitman 2 played well on the 3500X, certainly good enough, although the 1% lows were more successful than on a modern 8- and 12-thread processor. The overall performance was comparable to that of the Ryzen 5 2600, at least when testing with 1080p.
Even at 1440p, the weaker than expected low power of 1% persists, and although the average frame rates are similar to the 2600 and 2600X 2nd generation, the minimum frame rates are much lower.
Finally we have Total War: Three Kingdoms and here the 3500X is comparable to the first generation Ryzen 5 1600X, so the 3600 was much better at 1% low numbers. Here the 3600 was almost 40% faster, which is a massive increase in performance thanks to the use of SMT.
As soon as we switch to 1440p, where the GPU is the most important factor, all edges are removed, and here the 3500X works like any other CPU in our diagram. Only the R5 1600 falls a bit off the pace here.
Is it worth?
These are all the blue bar charts that we (almost!) Have for you. Now is the time to decide if buying the Ryzen 5 3500X is worth it if you come across it. For those who buy today, the best price we found online is $ 155. As already mentioned, it is about 20% cheaper than the R5 3600 in the USA. For other Aussies, it's about 30% cheaper.
Depending on the prices in your region for first and especially second generation Ryzen, the 3500X may or may not be worth it. The Ryzen 5 2600 is available in the US for just $ 120, and frankly it's a much better deal at this price, regardless of what you do with your system.
For productivity tasks, the R5 2600 is clearly the better choice because it was faster than the 3500X on every single core-heavy workload we did. When you combine the better performance with the lower price, you have an obvious winner. Even when it comes to gaming, the superior IPC of the Zen 2 cores of the 3500X can't make up for the lack of SMT support. To better illustrate this, here are some updated cost per frame charts.
Here, when you look at the frame rate performance over an average of 7 games, we see that the 3500X could only keep up with the second generation Ryzen 5 2600. They want to spend 6% more per frame, making it a poorer choice for US and Australian players at the moment.
If you only had the choice between a 6-core / 6-thread processor for around $ 150, the 3500X or the 9400F, we would choose Intel every time. Since there is the Ryzen 5 2600, this is the obvious choice.
For those who are concerned about gaming performance not just today, but in a year or two, the 3500X makes even less sense. If we look at 1% low performance in the 7 games tested, you get a good idea of how limited this CPU will ultimately be.
Would the 3500X be worth buying without more first or second generation Ryzen parts with more threads? If you play mainly, we recommend the Core i5-9400F instead, while you can choose either option for productivity reasons. The real scenario is that AMD Ryzen outsources 3600 processors that have not passed quality assurance to OEMs. This may make sense for prefabricated budget PCs.
For off-the-shelf purchases, the Core i5-9400F seems to be the better choice at $ 150, given the warranty issues that can occur with the 3500X. As a 6-core / 6-thread processor, the 9400F is probably not the smartest investment either. The currently best budget CPU is the Ryzen 5 2600 or Ryzen 5 2400G (if you need integrated graphics) or the additional $ 50 for the Ryzen 5 3600, an amazing CPU for the money that will undoubtedly keep years coming.