• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

R3 1200 paired with a Vega 56/64, bad idea or fine?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Vishera

Member
Joined
Jul 7, 2013
So, we're expecting Ryzen 3000 next year. I HIGHLY doubt it'll be at CES, probably not before Computex actually, but that's personal conjecture. But, suffice to say, it's likely coming some time in the next 6 months or so. With that in mind, I don't wanna upgrade my 1200 to a 1600 or 2600 with Zen 2 so close. Besides, I believe Zen 2 will be a much more worthwhile upgrade than Zen+. If I can get a 3600 in a few months' time, why buy a 1600 or 2600 now? So I'm upgrading my GPU instead, to either a Vega 56 or 64. My question is, at 1080p ultra, shooting for 144 FPS in most popular games (CS:GO, GTA V, the newest Assassin's Creed games, Fortnite, Shadow of the Tomb Raider, etc) will the 1200 massively bottleneck these GPUs? If so, by how much?

If I'm looking at taking myself from a potential 144FPS or more with a 56/64 plus a 1600, to like, 120FPS with a 1200, I'm not too worried about that. But if It's gonna hold me back by a significant margin, say not being able to break 100FPS...then I'll just drop the $150 on a 1600 and eat the loss when Zen 2 comes out. Not that big of a deal. I just wanted to consult the ever wise gurus of tech. Thanks, mates. :)
 
It’s often a bit of a misconception that the lower end processors are not good for gaming. Many games do not utilise all the cores available anyway so having two less cores than a ryzen 5 1600 will make very little difference and not worth splashing the cash to buy one of those.

I do not have much idea of what to expect myself but I just googled and an article from tweaktown came up which used an r3 1300x and a vega 56 and got over 110 FPS in tomb raider at 1080p

That processor is a little faster than yours so results will vary but you can always overclock to get some extra speed.

But yeh don’t worry about upgrading your cpu at this moment in time if you are waiting for the 3rd gen ryzen’s to come out.


 
Zen 2 will be probably presented at CES, then will be announced premiere for maybe March/April, then will be typical delay or early product problems so can expect it will be actually wider available in May/June.

I just say keep the R3 1200 till new chips will be released as we don't know if we have to wait 2-3 months or 4-6. I guess you can live till then on a slower CPU which is still good enough for games.
 
The difference will depend on the game and you'll have to try and dig through other benchmarks to see how the CPU impacts the highest frame rate. 4c4t is by current standards on the low end and while I can't say how much, it isn't going to be ideal for hitting highest frame rates.

However, is the system in use the one in the sig? In particular, I note 75 Hz monitor. I don't know what AMD's fast sync equivalent is called, but outside of that scenario isn't the ideal goal simply not to drop below 75 fps? With a GPU upgrade you can turn on all the eye candy for the best quality at that rate. With a CPU upgrade, is it really going to help unless you also update monitor?

On above assumptions, I'd hold off on the CPU upgrade and wait it out unless you're hurting too much already.

I'm running other benchmarks at the moment, but I could "simulate" a 1200 with my 2600 system with Vega 56 in it. The only game I have on the list is GTAV though. I think I saw that sottr now has a demo out so depending on how that runs, I might be able to get some feel on that also.
 
The difference will depend on the game and you'll have to try and dig through other benchmarks to see how the CPU impacts the highest frame rate. 4c4t is by current standards on the low end and while I can't say how much, it isn't going to be ideal for hitting highest frame rates.

However, is the system in use the one in the sig? In particular, I note 75 Hz monitor. I don't know what AMD's fast sync equivalent is called, but outside of that scenario isn't the ideal goal simply not to drop below 75 fps? With a GPU upgrade you can turn on all the eye candy for the best quality at that rate. With a CPU upgrade, is it really going to help unless you also update monitor?

On above assumptions, I'd hold off on the CPU upgrade and wait it out unless you're hurting too much already.

I'm running other benchmarks at the moment, but I could "simulate" a 1200 with my 2600 system with Vega 56 in it. The only game I have on the list is GTAV though. I think I saw that sottr now has a demo out so depending on how that runs, I might be able to get some feel on that also.

I'll be upgrading to a 144Hz display to replace the Emerson TV when I grab the GPU. If you could run a few tests for me, I'd appreciate it. :D
 
As was said, it depends on the title as to how much, if any, performance loss you will see. 1st gen ryzen does tend to hold higher end+ gpus back at 1080p. That said, I'd get your GPU and monitor and wait for your 3 series. ;)
 
I'll be upgrading to a 144Hz display to replace the Emerson TV when I grab the GPU. If you could run a few tests for me, I'd appreciate it. :D

I just downloaded the sottr demo and it does include a built in benchmark. Updating my AMD system software and drivers right now but will run it later, when I will also revisit the current OC challenge and hopefully take 2nd place at least.

Do you overclock the CPU? It is easy enough to limit it to 4c4t (I believe in 2+2 config), but at what clock? I note it is listed as 3.1 base 3.4 boost, +a tiny bit xfr. Unless I hear otherwise I'm tempted to run at 3.6 fixed as a balance of an easy OC without needing higher end cooling.
 
Based on the above descriptions, I will be testing:
1, Ryzen 2600 + Vega 56
2, Simulated 1200 + Vega 56 (reduce cores to 2+2, turn off SMT, undecided clock)

GTAV and SOTTR have built in benchmarks and I'll be using that at 1080p with high-ish settings. I don't know how much difference it makes but SOTTR recommend "high" as default on my 1080Ti + 8086k system, with "highest" setting available above that. 1440p "highest" gave me average 76 fps 75% GPU bound, and 1080p "highest" gave me 103fps 35% GPU bound. So for 1080p it seems to suggest even a stock 8086k is the limiting factor. I'm not complaining at being limited to 100fps...
 
20181209152129_1.jpg
2600 stock, Vega 56, 1080p high

20181209152839_1.jpg
2600 stock, Vega 56, 1080p highest

20181209164052_1.jpg
2+2c 3.6 GHz SMT off, Vega 56, 1080p high

20181209171118_1.jpg
2+2c 3.6 GHz SMT off, Vega 56, 1080p highest


GTAV built in bench

Frames Per Second (Higher is better) Min, Max, Avg
Pass 0, 13.368572, 128.445557, 102.174431
Pass 1, 75.628090, 166.850189, 127.958229
Pass 2, 45.939415, 166.038483, 123.775757
Pass 3, 61.436760, 220.541214, 141.846619
Pass 4, 37.234932, 255.167114, 134.279053
2600 stock, Vega 56, 1080p defaults + vsync off

Frames Per Second (Higher is better) Min, Max, Avg
Pass 0, 36.134998, 108.135002, 86.622734
Pass 1, 73.151817, 139.025986, 108.054588
Pass 2, 74.501213, 175.632706, 106.539619
Pass 3, 45.054180, 156.467590, 111.185013
Pass 4, 4.964753, 230.680496, 114.041985
2+2c 3.6 GHz SMT off, Vega 56, 1080p defaults + vsync off

=== SETTINGS ===
Display: 1920x1080 (FullScreen) @ 59Hz VSync OFF
Tessellation: 2
LodScale: 1.000000
PedLodBias: 0.200000
VehicleLodBias: 0.000000
ShadowQuality: 2
ReflectionQuality: 2
ReflectionMSAA: 0
SSAO: 2
AnisotropicFiltering: 16
MSAA: 0
MSAAFragments: 0
MSAAQuality: 0
SamplingMode: 0
TextureQuality: 2
ParticleQuality: 1
WaterQuality: 1
GrassQuality: 0
ShaderQuality: 1
Shadow_SoftShadows: 1
UltraShadows_Enabled: false
Shadow_ParticleShadows: true
Shadow_Distance: 1.000000
Shadow_LongShadows: false
Shadow_SplitZStart: 0.930000
Shadow_SplitZEnd: 0.890000
Shadow_aircraftExpWeight: 0.990000
Shadow_DisableScreenSizeCheck: false
Reflection_MipBlur: true
FXAA_Enabled: true
TXAA_Enabled: false
Lighting_FogVolumes: true
Shader_SSA: true
DX_Version: 2
CityDensity: 1.000000
PedVarietyMultiplier: 1.000000
VehicleVarietyMultiplier: 1.000000
PostFX: 2
DoF: true
HdStreamingInFlight: false
MaxLodScale: 0.000000
MotionBlurStrength: 0.000000
 
So not terrible. But not 144FPS either. I wonder how much that would be affected by me going with a 64 instead...food for thought I guess. Thanks for the help, Mack! :)
 
Will the monitor be FreeSync? I've found G-sync on my main system to help a lot with the feel, that reaching 144 fps isn't a goal I aim for any more, as long as it is somewhat above 60 it's all good. I don't play the fastest action of games though, none of the twitch e-sport types.

Also, forgot to say, SOTTR bench also writes out a text file with frametimes in it, in case anyone really wants to analyse it beyond that presented on screen.
 
Yes, will be getting a FreeSync monitor. I'll have to try some different combinations of hardware and graphics settings to see what works best.
 
Back