Graphic Card under $200

mrethan

Reputable
Nov 23, 2015
34
0
4,530
Hey Guys! 😀

I've been looking for a decent graphics card for the price under $200.
I've heard lots about the GTX 950/960 and other AMD R9 stuff.
I'm not sure what to make my mind up on, I was wondering what you guys would be thinking.
If you do suggest something please make sure it supports my motherboard and such

SPECS
-----------------------
Motherboard: MSI 760GM-P34
Processor: AMD FX-6300 3.5 ghz
RAM: 8GB
Graphics: AMD Radeon R7 200
 
Well, actually, I do not believe there is any difference at all. The only game that improves fps with the 4GB version I think is AC Unity, which simply poorly allocates the amout of VRAM available on 2GB GPU's, aka It's just poorly optimised. But I thought for 200 dollars perhaps it could prove to be the best option.

I believe, other than that anyway, that the GTX 960 vs R9 380 thing is all about PREFERENCE. The performance is slightly better on the R9 380, but I prefer the efficiency, drivers and features of the GTX 960 (My GPU).

Video Card: MSI Radeon R9 380 2GB Video Card ($185.99 @ NCIX US)
Total: $185.99
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2015-12-28 06:37 EST-0500

Video Card: EVGA GeForce GTX 960 2GB SuperSC ACX 2.0+ Video Card ($194.99 @ SuperBiiz)
Total: $194.99
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2015-12-28 06:39 EST-0500
 

So what do you believe is the best GPU out of the ones you picked?
 


Alright, I'll probably pick that up.
 


It is proven that even if a game uses 3 to 4GB VRAM, the fps and consistency of it is still just like the 2GB version compared to the 4GB version. I don't have a clue in what way somehow the 4GB version is going to perform better to be fair.

So far from every benchmark that I've researched, let it be Google or even Youtube, the fps is always the same on average. And this is from new games that use way more than 2gb vram.
 


From Digital Foundry;
This all translates into a healthy state of affairs for the R9 380 during the all-important 1080p benchmark run, where we see AMD's latest match or exceed its Nvidia competition in all but one game - Battlefield 4. But what is especially of interest are the results on Ryse, Assassin's Creed Unity and Far Cry 4, where the 4GB version of the R9 380 produces noticeable improvements of the 2GB card. In the table below, the 2fps uplift may not seem like a big deal, but we would strongly suggest that you watch the video to see the difference visualised - stutter in each of those titles is considerably reduced with 4GB on tap. It's also no surprise to see that the older 3GB R9 280 also does well, presumably owing to its additional gig of VRAM over the 2GB cards. The bottom line is clear: these games stutter less with more VRAM, and it seems that 2GB isn't enough to house top-end texture quality on certain games.

[video="https://www.youtube.com/watch?v=ESeFXwGfBsI"][/video]

http://www.eurogamer.net/articles/digitalfoundry-2015-radeon-r9-380-2gb-4gb-review
 
I'm not going to post against that, it seems accurate. But what I'm wondering, which is suprising, is that the 4GB models accelerate in games that support up to 8 cores/threads, like Far Cry 4 and Battlefield 4. Now based on the research I've done is that frametimings are far better on AMD FX 8xxx processors, compared to the i7-4790k's virtual hyperthreading (which produces more frames, but is often criticised by enthusiasts because they believe Hyper threading is not a good asset for real-time rendering in a game that uses up to 8 cores/threads) in games like BF4 that support up to 8 cores.

So now I am wondering, in said games where up to 8 cores are supported, how the R9 380 2GB vs 4GB would run in an FX-8320 based system. I think it should either show a more consistent but lower amount of fps on the 2GB model, or it should consistently be closer to the R9 380 4GB without stuttering.

And believe it or not, if the majority of games supported up to 8 cores, I would be running an FX-8320 system now. So I'm really wondering if the same inconsistency is also present when testing it that way.

But It's clear anyway, the R9 380 4GB is more consistent in CPU intensive games on Intel's processors.
 


It's only something I've recently done my research about, after it was proven to me that FX-8320 is far better than I thought they were when used in a properly chosen system with a good 990FX motherboard etc.

Apparently, what I've been told but haven't done enough research about, is that Intel hyper threading is a technique based from the 80s that was reintroduced recently with obviously Intel's i3 and i7 processors and has a negative side of it for real-time rendering. Basically virtual cores are not close to real cores, this feature is higher rated than it should be basically.

Also 990FX mobo's are better supported for SLI/CF configurations compared to Z97 and even Z170, because they still have a lot more PCIE lanes in capacity (42) compared to 16 PCIE lanes on Intel's latest Z170 motherboards. This is also something I didn't know atleast.
 


My first response would be that even though AMD is the sponsored brand in BF4, that Nvidia actually did perform equally UNTIL October 2015. Since then Nvidia GPU's took a hit on performance and now drop a lot more in fps. I'm finding the situation very fishy. But that video is ofcourse Pre-October 2015, so performance is equally good when comparing GTX 960 vs R9 380 in that video.

Also AMD is heavily better supported in Battlefront compared to Nvidia, atleast this was the case when the game was just released. You got 10-15 fps more with an R9 290X compared to GTX 970. Since that game is released you also get less fps with Nvidia cards in BF4, hence why It's fishy for me.

I get your point though, I mean even though the R9 380 2GB seems to dip it is ONLY a fraction compared to the 4GB model in Battlefield 4. I guess that only so much an increased amount of VRAM can do, since the rest of the specs I think are still exactly the same?
 


Yeah, that kind of debunks the whole "4GB is required" mantra doesn't it?
 
I recommend the GTX 960 simply because of Nvidia Experience (NE). It makes it easy to install/update drivers, and it will set graphics settings on many popular games. Some people don't like NE, but it does make the things easier. There's also built-in support for game capture and streaming if you're into that.
 
Kinda, the 2GB models are always going to be better price/performance rated anyways. I wouldn't quickly recommend a 4GB model unless It's very well priced compared to the 2GB models. But often the 4GB models are overrated by the performance aspect by the majority compared to how they actually perform vs. the 2GB models, no question about that.



AMD also has that (Game DVR), I've used it. But I find Shadowplay better. But It's also the frequent driver updates, the 3D settings that you can change. From what I know and experienced this doesn't work as good with AMD (MFAA, Enabling FXAA through Nvidia's panel, Adaptive V-Sync also through Nvidia control panel).

For those things you're going to need an alternative program or a work-around to make adaptive V-sync work, which is something I often use.
 

The point is that if the difference is already noticeable right now, it's only gonna get worse going forward. Maybe 'required' is too strong a word.

Yeah. I'm still salty that I went with the 1GB version of my HD 6850 card rather than the 2GB version and I'm currently memory limited rather than GPU limited, and anyone that goes with a 2GB version when there's a 4GB version available, will experience that same thing I did within two years or less. Being memory-limited is not fun, at all. More memory is better for the longevity of your card if it's mid-range or higher.

And what I posted was just one example. The same is true for the GTX 960;

960-4v2gn-acu.jpg


http://www.gamersnexus.net/guides/1888-evga-supersc-4gb-960-benchmark-vs-2gb/Page-2
 
The AC Unity game is known for poorly allocating the VRAM available of GPU's at 2GB and lower though. I would not use it as an example although I understand the point you're trying to make. Unity is very poorly coded.

Also in that benchmark, we already discussed this with some others before, is that they used a higher clocked GTX 960 4GB than the GTX 960 2GB, that's why the fps was higher on the 4GB model, rather than because of 4GB VRAM. Stupid decision by them.
 
With an fx chip & that corsair PSU the 960 is a better fit for your system IMO.
Performance difference between the 960 & 380 is negatable, its proven to be even less paired with an amd CPU - the 960 manages more draw calls under dx11 than a 380 when paired with an and system - its strange but true that theoretically the nvidia cards pairs better with amd than a comparable amd GPU does!

The 4gb vs 2gb is again debatable , there are 4gb cards available for $207 , at that price point its probably worth going for a 4gb card just for the safety factor
 

Even though you are correct, the difference in lowest FPS is still remarkable. 5 fps vs 30 fps is not going to be due to any difference in clock speeds. We know we can't count on properly coded PC games all the time nowadays either.

It's too bad they didn't show more benchmarks with the minimum framerate. Averages can show up fine, but the drops are what's annoying. My HD6850 can handle Dragon Age inqusition fine at a steady framerate. Turn the camera and I drop to 1 fps for a short while since the memory buffer needs to be re-filled. My average will still show as playable, even though it really isn't.
 
Good discussion though. The first reply in the other thread when I described AMD FX-8320 is an alternative option for rendering was ''Wow you are blind AMD FX-8320 is NOT a true 8-core CPU, Intel i7 is much better!''. The hate speech is real 😀.
 


So I should stick with the Radeon R9 380?