Following on from our recent revisit of the RX 570 and RX 580, I wanted to upgrade our Athlon budget build with a dedicated graphics card to see what kind of performance we’d achieve compared with the same RX 570 in a much higher-end platform. This would effectively demonstrate how much performance you can realistically expect by staying with a low-spec core platform like the ASRock A320M-HDV and AMD Athlon 200GE APU.
First of all, I want to acknowledge both the ASRock A320M-HDV motherboard and AMD Athlon 200GE APU – these products have made owning an entry-level PC affordable to many that may not otherwise have been able to do so. The A320M-HDV can be had for around $80 and the Athlon 200GE was around $90 on release. If you’re interested in more detail about the low-cost platform, you can read the launch reviews of the Athlon and ASRock A320M-HDV I’ve linked here.
There are limitations of the Athlon process that I’ll highlight in this article but these should not be seen as a criticism or negative option of the product itself, they are simply the capabilities of the entry-level technology. The AMD Athlon with Vega 3 graphics has delivered performance and value that we rarely see in the entry-level processor or APU tier.
Whilst the Athlon 200GE is a ‘locked APU’ and the A320 chipset doesn’t support overclocking either, this core platform delivers a dual-core CPU with SMT for 4 logical processor threads running at 3.2GHz, 3 cores of Vega Graphics, PCIe 8x, an effective stock cooler, an M.2 slot and an HDMI port. Add some RAM and a cheap SSD and you’re good to go. Surprisingly, the stock cooler isn’t loud and the ‘upgraded’ system with the Sapphire RX 570 Pulse installed remained quiet despite the punishment we gave it.
The key concerns from a gaming perspective are multi-core performance and clock speed. What this means in real terms is that the Athlon 200GE can struggle to do more than one thing at a time. As an example, if you are playing a game and Steam tries to download an update or your antivirus scan kicks in, that game can turn into a slideshow whilst the CPU diverts its attention to the background tasks. The Athlon 200GE also has VEGA 3 graphics which also means that gaming is possible but will typically be limited to 720p resolution and low settings. In most cases – whilst this is better than what you can expect from similar-era Intel Pentium or i3 processors – it’s not really gaming-grade.
By being careful with background task management, you can mitigate the lack of cores of the Athlon CPU to a point, but the best way to improve frame rates to a respectable level is to add a dedicated graphics card. Keeping in mind that our Athlon test platform without a discrete graphics card is worth roughly AUD$600 and that the APU/CPU will present a bottleneck, we wanted to look at how well the AMD RX 570 performed as a replacement for the embedded VEGA 3 graphics. The RX 570 can still be purchased new for around AUD$220 and it’s regarded as a good value 1080p gaming card for those on a budget. The price makes it a sensible consideration for existing Athlon 200GE owners who are considering a graphics upgrade.
The RX 570 that I’ve tested here is the Sapphire Pulse 4GB OC. It’s a factory overclocked RX 570 with a quiet but effective after-market cooler and retails for AUD$220.
In the benchmarks, you will see a direct comparison of the RX 570 in both an A320M/Athlon 200GE platform and our standard ASUS ROG STRIX X470-F Ryzen 2700X test bench. To be clear, the ASUS ROG STRIX test bench should win in every benchmark due to faster RAM speeds, PCIE x16 lanes (compared to the x8 bandwidth of the Athlon 200GE), 8 cores and faster clock speed of the 2700X CPU. Make no mistake, this isn’t a fair fight but I wanted to demonstrate the RX 570 under ideal operating conditions and the compromise that you’ll make if you add it to the lowest-spec (but still viable) PC that I’d build at the moment.
For reference, the specifications of each test system are listed side by side below.
|Althon 200GE Rig||ASUS ROG Test Bench|
|AMD Athlon 200GE|
|AMD Ryzen 7 2700X
|16GB 2666MHz ADATA Dazzle (2x8GB)||32GB 2933MHz HyperX Predator RGB (4x8GB)|
|ASRock A320M-HDV Motherboard||ASUS ROG STRIX X470-F Gaming Motherboard|
|256GB ADATA SX900 SSD||Samsung Evo 840 250GB SSD
ADATA SX7000 M.2 NVMe 256GB SSD (Boot)
|Seagate 2TB Firecuda 3.5" SSHDD||WD Black 6TB 3.5″ HDD|
|Corsair AX850 PSU||Thermaltake TOUGHPOWER I-series RGB 850W PSU|
|Sapphire Pulse RX570 4GB OC||Sapphire Pulse RX570 4GB OC|
In these graphs we see the physics performance of the Ryzen 7 2700X give a boost to the overall score compared to the lesser Athlon as expected but the graphics performance when compared in isolation isn’t that far off.
The gap was at it’s closest in the FireStrike Ultra test as the RX 570 4GB was getting hit hard and the difference in CPU wasn’t as much of a factor in the scores.
In the FireStrike and FireStrike Extreme benchmarks above, the graphics score is only slightly ahead but the overall score is more significantly influenced by the extra power of the RYZEN 7 2700X CPU.
Sky Diver is the least taxing benchmark we run and the overall score of the Athlon system is almost half of the 2700X despite the isolated graphics card performance being only slightly down.
Time Spy is the DX12 benchmark that also weights the CPU performance higher which explains the big lead in overall score, but again, we see only a minimal lead in graphics performance.
The Futuremark benchmarks are demonstrating a small hit to graphics performance on the Athlon platform but it isn’t really impacting the RX 570’s isolated scores. What’s a little worrying is the overall scores due to the CPU performance but we’ll cover that in some in-game benchmarks and real-world gaming performance shortly to see how much it matters in the real world.
The Unigine benchmarks don’t break down the CPU and GPU elements, we just receive an overall score, but the tests are generally more taxing on the GPU so these are still an interesting indicator.
The Unigine benchmarks show how the bottleneck moves from the CPU at lower resolutions to the GPU at higher resolutions. These results are consistent in terms of a pattern across all three benchmarks.
These benchmark results are generated from benchmark functions built into the games and not subjective or prone to variables as we’d see in a play through.
Far Cry 5
I’d been asked about this one by a few readers so I’ve tested it at 1080p and 1440p at various settings to show the trade-off.
As you can see, the performance gap closes a bit at the higher 2560×1440 resolution but at 1920×1080 there is some significant performance being left on the table.
The game is playable and the CPU load can get high when there is a lot going on, but overall the performance and experience wasn’t too bad at 1080p with High settings and FreeSync enabled.
Tom Clancy’s The Division 2
The benchmarks don’t tell the full story here but I’m going to include them anyway. This is one of our standard in-game benchmark tests.
The 1080p benchmarks look playable at medium or even high settings. At times, it played ok too. Based on these results one could question the validity of the minimum specs – except that when running this game, any activity in the background at all can reduce it to a slideshow or even crash it out completely. This title can be CPU-intensive as it loads the map constantly while you move around and the dual cores don’t seem able to keep up despite the multi-threading.
For those people that emailed me about this one on an Athlon and asked for the benchmarks – here are the benchmarks, but the developer meant what they said with minimum specs and the Athlon 200GE isn’t it. I did give it an “A” for effort, though!
Shadow of the Tomb Raider
This is another standard in-game benchmark that pushes graphics cards. It’s in the standard set so I’ve included it below at 1080p and 1440p.
The Athlon 200GE was unable to achieve 60FPS at 1080p on low where the performance difference was at its greatest to the 2700X system.
Rainbow Six Siege
Up to this point, it’s been a rough start for the Athlon but things turn around in this title.
Rainbow Six Siege does well here and the frame rates are good enough to really get into the game. Sure, the performance gap between the Athlon and the much more expensive 2700X is significant but when you look at what the Athlon can deliver in this scenario, it’s still over 120FPS in the benchmark at 1920×1080 at the Very High preset.
Whilst these results are impressive, remember that I’m not running any game-launchers or background processes. The Athlon has been allowed to focus those two cores and four threads on simply playing R6 Siege.
This one is an “oldie but a goodie”, it isn’t that demanding, is a good game to spend some time playing and it’s often available on sale. Perfect for a system like the Athlon!
The benchmark gave over 110FPS at 1080p and 80FPS at 1440p – I included 4K because the results were so close. The performance of the low spec Athlon rig in this test is pretty good for the money.
This title is a tough DX12 taskmaster. The settings used are listed below.
The difference in CPU is actually the difference between an playable experience at Ultra and not playable at all. If you tweak the settings there are some gains, but overall this is another title that has minimum specs for a reason – even at 1080p.
I’ll keep including this until GTA VI comes out and also because the benchmark is very consistent. The settings are below and these could be tweaked for better performance but we have a standard and this is it for our testing.
These results are encouraging and gameplay was pretty good for the price. At 1080p with FreeSync, it’s not bad, but with a few minor tweaks to the quality settings, 60FPS average isn’t hard to get.
In the interests of being thorough, I tested the titles below using MSI Afterburner and monitoring the FPS counter and CPU utilisation. I played at 1080p given that this is the most likely resolution I’d expect an Althon 200GE system owner to be using. I also deliberately chose the titles based on reader requests.
I tested this at 1080p with the settings on High as I’d baselined the RX 570 with these for the RX 570/RX 580 article in March using a RYZEN 2700X. It’s also the resolution I’d expect a low spec/low-cost gamer to be targeting.
The Athlon 200GE / RX 570 combination served up 110FPS at 1080p with max settings and this was very playable. Given the age of this title and its reputation for having modest requirements I’d expected it to be a good experience – it didn’t disappoint.
With the settings on High and 1080p, I saw a frame rate range of 45-60FPS and the game-play was smooth enough, averaging about 55. CPU usage was around the 70% mark at times but would regularly drop to around 50% so the Athlon 200GE didn’t seem to be struggling too much. When loading games, the CPU worked a lot harder. The key here is to make sure that you have nothing else running in the background. I had a background system task startup at one point and the game became sluggish while the Athlon sorted out its priorities.
In comparison, the experience on the high-end gaming platform with a RYZEN 7 2700X processor was generally 53-67 FPS with a typical average frame rate of 60 FPS.
Overall, this is a good result for the Athlon platform if you can manage background tasks.
This isn’t in our typical testing routine but I’ve had people asking about the Athlon as a cheap second PC for Fortnite.
I started with Epic Quality settings at 1080p and noticed straight away that the poor Athlon 200GE was getting smashed by Fortnite and reporting 100% CPU usage in MSI Afterburner. This dropped down to the high 80s at times when there was nothing going on around me. As soon as I’d get to the smaller circle and top 10 players remaining, the lag made this a frustrating experience. Dropping the quality to the High preset still had the CPU regularly hitting 100% but it tended to spend more time at 95% and there was far less lag at the busy times.
My experience with High settings at 1080p was 60-69FPS; enjoyable enough but if you are hardcore when it comes to Fortnite, I’d suggest stretching the budget and buying a higher spec CPU with more cores and a faster clock speed.
Using Ultra settings at 1080p, I saw an average of around 100FPS on the practice range but the CPU usage was at 100% for the Overwatch application.
By locking the FPS to 60 I saw a reduction of CPU usage down to a typical 88% which made the game-play smoother. At a locked lower frame rate, Overwatch played just fine without any noticeable stutters once the level had fully loaded.
I used the training mission on the High preset at 1080p for this run as these settings previously delivered 75-85FPS on our 2700X/RX 570 platform. The frame rate was about 50FPS with the Athlon but it should be noted that it was also much more variable depending on what was happening in the game and in any background or system tasks.
The best performance I saw was on the practice range with no other players or NPCs/bots. As soon as the CPU has to work hard the frame rate suffers.
This one is an old favourite and the Athlon 200GE with an RX 570 gave us a typical range of 80-110FPS with an average of 90 overall when set to max quality and 1080p resolution.
Borderlands 2 is often on sale for a bargain and goes well with a low-cost platform like this one.
Based on the synthetic benchmarks, the raw GPU performance in isolation isn’t hit too hard by the budget platform, but the processing power of the CPU to handle game instructions, physics, any decompression of game files on the fly and general compute for other entities in the game world obviously holds the Athlon back as a gaming option. The argument of ‘single-threaded games should be ok’ that I’ve seen on forums probably needs a caveat noting the impact of innocuous background tasks that also compete for the CPU’s attention.
Should you upgrade your Athlon 200GE (or similar) with an RX 570? It’s probably going to be worth it for the most part but it will ultimately depend on what you want to play. Check the specs and be prepared to drop a new quad-core CPU in if you can’t live with frame drops. The Ryzen 5 1500X or 1600 might be worth a look if you can find one secondhand.
As an example, I took a screenshot of the Radeon Software ‘Gaming’ tab with the status listed for the installed games. Most are marginal or have an alert that the system doesn’t meet the minimum specification. Whilst this doesn’t mean you can’t play them, it does indicate that your experience will likely be lower than what the developer intended as the base level.
CPU intensive games and games optimised for multi-core are likely to struggle with the dual-core Athlon and managing background tasks is essential. The Witcher 3 and Battlefield V are good examples of game-play suffering with anything else running in the background. This also goes to show why games have a minimum specification and what happens when you don’t meet it. You should expect stuttering in titles like this – Witcher 3 wasn’t that bad but I considered it to be frustrating in Battlefield V.
Older games and games developed for use as single-threaded applications may run just fine. The performance of CS:GO, Rainbow Six Siege, BIOSHOCK Infinite, Overwatch and Borderlands 2 were solid experiences and highlighted what I’d consider to be the bonus value of the Athlon as a CPU and the A320 platform.
Overall, although I was impressed with the experience of the RX 570 paired with our Athlon build for the money, I wouldn’t necessarily recommend it as a general rule and certainly not as a new build in 2020. If you have an Athlon (200GE or greater) then this article shows you what you can get with a lower-end graphics card like the RX570 and where the bottlenecks exist. If you’re building a new budget gaming rig, I’d suggest stretching to a RYZEN 5 second-generation or perhaps the upcoming Ryzen 3 due in June 2020 if you can stretch for that extra grunt.