GTX 770 vs Radeon HD 7970 vs R9 280x

Page 5 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.

GamerMT

Honorable
Sep 27, 2013
49
0
10,530
Hey. I have a GTX 650 and i can't really choose between this 3 cards. Games that i play are:
Battlefield 4
Neverwinter
and Warface.
But i'm getting a new card because of Battlefield 4. This are my PC specs:
CPU: AMD Phenom II X4 965
GPU (currently) Nvidia GTX 650
RAM: 8GB RAM DDR3
HDD: 1TB
OS: Windows 7 32 bit (getting 64 bit of course)
so.. which one should i get? 7970 will be MSI Lightning Boost edition or CLUB 3D , R9 280x will be Toxic or CLUB 3D, and GTX 770 will be MSI Twin Frozr OC or Gigabyte OC. Thanks guys!
 
Solution
People should try to be a little less emotional about this debate.

You can show me all the synthetic benches and FPS tests you want, but my real life experience using both GTX 780, 770 and 7970 Vapor X, etc, is that the Nvidia cards are more reliable higher performing over time. I've had all kinds of driver and reliability problems with the AMD cards, but literally zero issues with an EVGA 780 FTW and MSI Gaming 770s. For a long time I was a AMD fan and used them exclusively in all of my personal and business builds. But as the reliability issues of AMD both in terms of inconsistent performance and poor driver support became problematic over a number of months in a number of different builds, and I didn't have any such issues with...
Hi guys :)

So I was checking out these bi-monthly updates on price to performance, and they lead me to choose the 770. I was planning on getting this overclocked (Core Clock 1150 MHz, Boost Clock 1202 MHz, Effective Memory Clock 7200 MHz) one for $327.

Then I got a newegg email with a code that puts this 280x (Core Clock 850MHz, Boost Clock 1000MHz, Effective Memory Clock 6.0GHz) at $290. It's not overclocked though and I don't know how difficult it would be to try it myself.

Think I should go for the 280x and try to overclock it myself or will the overclocked 770 be significantly faster?
 

The pny 770 is going to fare much better than a lower tier xfx.
 
People should try to be a little less emotional about this debate.

You can show me all the synthetic benches and FPS tests you want, but my real life experience using both GTX 780, 770 and 7970 Vapor X, etc, is that the Nvidia cards are more reliable higher performing over time. I've had all kinds of driver and reliability problems with the AMD cards, but literally zero issues with an EVGA 780 FTW and MSI Gaming 770s. For a long time I was a AMD fan and used them exclusively in all of my personal and business builds. But as the reliability issues of AMD both in terms of inconsistent performance and poor driver support became problematic over a number of months in a number of different builds, and I didn't have any such issues with GTX 770+ cards, I just converted everything over to Nvidia. My 3 home gaming rigs now house MSI Gaming 770 and EVGA 780 FTW cards with my personal favorite definitely the FTW. At stock speeds it's only about 20% better than the 770, but they overclock really well and without even pushing it I'm now at a good 40% improvement over the 770s. The 7970 Vapor xs I've used perform similar to the 7970 in synthetic benches, but longevity and inconsistent driver issues were an issue with both cards I got. They also could not tolerate an overclock during heavy gaming. They would make it through the bench, but if you game for 2-3 hours then there were a lot of blue screens.

I could care less about the whole "fanboy" garbage - I base my decisions on real life experience, and the main things I look for are 1) consistent high performance 2) driver support and reliability and 3) card longevity. In my experience the AMD cards aren't even close to the GTX series, and that's before the upcoming Maxwell release. Note that I was "converted" based on real life experience with a multitude of cards and various core configurations over a number of years.
 
Solution


That seems quite emotional and subjective.

Yeah, nVidia driver support is unsurpassed; 60 pages of glowing accolades.
https://forums.geforce.com/default/topic/720880/geforce-drivers/official-nvidia-337-50-beta-performance-display-driver-feedback-thread-4-7-14-/1/

 
Have you guys see this?
900x900px-LL-47b12832_Titanfall_Benchmarks_Retail_20140324092345-pcgh_140324093721906.png
 
after having a 7950 for over a year, proud owner of a classy 780 now, i personally didn't have any problems with amd drivers. i did enjoy the performance increases which "felt" good.

that said, today, now, there are very little performance differences between a full gk104 or tahiti chip. they really really REALLY are so close together in performance, especially out the box, that its really personal preference and if you need the proprietary features that each card offers like physx mantel etc.

since the 770 and 280x are obviously not the same 680 and 7970ghz of early days that were not voltage locked, if i really had a choice, i would take a brand new early revision msi lightning 7970 easy(it also still holds the gk104/tahiti world record) as it will overclock much farther than any of the other 7970s or 680s. now, today, i would take a classified 770 if i had to be pushed.... but i own a classified 780 so its kinda a mute point for me. but this is purely look at the cards as how far are they capable of going. out the box, both are still very capable and many many sites will tell you the same, same game settings, same experience.
 


well the point is? there are million of user out there so there are bound to be problem be it directly from nvidia or from user setups. go to the other side of of camp and you should see same problem. but you got far better response from nvidia when you report a problem to them. this is the recent examples:

http://techreport.com/news/26333/nvidia-geforce-337-61-beta-hotfix-display-driver-released
 


Alright someone is drunk off their rocker if you believe this. I don't even care about the top half.

BUT A GTX660 WILL NEVER GET ALMOST 50FPS WHILE A 770 GETS 30!!!!

Pls get info from a reliable source instead of typing in Google "Which Graphics Card fastest?!?"

Thix :^)
 


It would help if you noticed there are three 770 entries. One is "vsync on".
 
All you nVidia guys need to do is post a video running BF4 on 64 player Shanghai server 1920 x 1080p, all Ultra settings and 4x msaa etc. The video will show the FPS, graph and video info at the top, and also verify the video settings. Cap max FPS at 75.

Then in the console type 'perfoverlay.framefilelogenable 1' to capture the data while playing. Before stopping the recording, turn it off or simply quit the game.

Download http://www.bytemedev.com/bf4-fta/ , import the data and post the results.

Like this:
1Vn50xH.png


And here's the video of my 280x running Mantle, capped at 75 FPS.
https://www.youtube.com/watch?v=NxRMNmyTxKo&feature=youtu.be

Forget the outdated benchmarks, it's real world user results that matter. The 280x stomps the 770 and I'll bet the 780 can't beat it.

 
I'm not sure you can say the 280x stomps a 770. Would that be according to the other charts compared to the chart above? This whole thread has turned into a pissing match. Quite entertaining I might add. I'm gonna try the real world results on a 780 see how things fair. Thanks slimething for posting useful information.
 
i ask myself at the moment which one i should choose.

Backgroud:
My graka is defect. Gtx580 Geforce. I never had problems with Geforce and i always used Geforce.

Now the problem is, that Nvidia technology/politics is always making money. Nothing wrong with it. Companies need money for new innovations. Sometimes the money is so important to some companies that they dont want to drop old technologie making more advertise and hoping that peoples buy it without questions.

I like NVidia much because of the support of gamers. But now with Mantle they are showing that they dont want the way of using standards, which would give the world a good base and making things not so complicated anymore.

So money is speaking. So yes Nvidia is a company that spend much money in prestige. AMD is a little bit different. They need money of course, but they dont say to good technologies that others cant have it.

With this way of handling things, AMD is more popular for rational thinking persons. But NVIDIA is making so much advertise and deals with others that they make people blind for the way it should be.

Dont get me wrong. Competition is always good. But sometimes companies should think what could bring the world to a next level on graphic applications "together".

Jeah, i talk about MANTLE. A perfect solution for a problem we have on the PC. But not a solution only for the PC. It is very mutable and open.

But because of money some companies dont want it but they cant ignore it too and the benefits.

So i dont know if it is personal, but the performance of AMD is much better on there reference model and the politic is much better.

But anyway as Battlefield 4 players you should buy AMD. Dice and AMD are having a partnership because of Mantle. So i will leave NVIDIA and go to AMD. More power and better chipset.
 


He uses google maybe, but "PC Games" is a popular magazin in my country. When we cant trust the "PC Games" anymore, who we could trust ^^.

So pelase dont write something like this without using Google once.

:^)

 

Yes, I'm saying for BF4 the 280x mops the floor with the 770. It should be very simple to prove me wrong, but thus far all I've seen is flapping jaws link wars.

If the 770 I had in November would have performed like this I'd still have it:
http://www.hardocp.com/article/2013/12/02/sapphire_toxic_r9_280x_video_card_review/4#.U1rEyaIvl8M
image.html


But it performed like this, so I returned it for the 280x:
http://www.hardocp.com/article/2013/11/17/battlefield_4_video_card_performance_iq_review/5#.U1rEK6Ivl8M
image.html


I actually said not to buy an AMD card because of the gpu miners artificially inflating the prices. In November I paid $300 for a 280x. At the time I thought it was the best bang for the buck, especially for BF4 and even more so if Mantle were to work out.

Up until 14.4 beta drivers, I've consistently stated Mantle isn't ready for prime time, but now Mantle is becoming more refined and the drivers are much improved overall, at least by my reckoning. I also suspect DICE has done some work in the patches to improve Mantle performance. AMD must get it right or will lose a lot of customers, myself included.

Yes, it has become a pissing match and it really shouldn't be, but after reading countless numbers of posts about how much superior nVidia cards are for BF4 and that Mantle is just hype I decided to throw down the gauntlet. In reality I couldn't care less, but as they say money talks, BS walks. Neither my card or cpu are overclocked; the card's voltage is locked anyway.

How about no more outdated 2, 3 or 6 month old "benchmarks", synthetic or otherwise, and no more single player meaningless FPS "tests". Multiplayer is where it's at.

What driver version are you running on your 780? It will be interesting to see the results.
 


Well that's interesting...

In that first link (December 02, 2013) for the sapphire toxic review, the cheaper stock 770 actually beats all the cards... Then in the second link (November 17, 2013) it gets 10fps less.

Looks like a driver update raised the 770's performance in bf4.

 


Ok Results. 780 stock clocks. CPU is stock as well. Very interesting.

05lf.png
[/URL][/img]
 




In console, type gametime.maxvariablefps 75 or whatever number you wish to cap it. Or, download BF4 Editor and have control over several settings at your fingertips.
 


AMD said they will 'open' Mantle but as long as AMD is the sole company that control the future of Mantle other graphic vendor will not going to adopt Mantle. the clue is clear. if AMD really is serious about making Mantle as a standard that is open for others to use they will invite every graphic vendor (intel, nvidia, qualcomm, imagination technologies, ARM, Vivante to name a few) to develop mantle together when they start developing for it. why only talking about it after two years in development? and other graphic vendor won't get to peek details on Mantle later this year or early next year when AMD hardware already able to run game based in Mantle right now. during Toms Hardware interview with Dice's Johan Andersson he reveals that Mantle will not work on hardware prior to GCN because they lack some hardware that needed by Mantle. and people from AMD already mention a few times that Mantle is build specifically for GCN hardware. that's mean even if AMD open the details about Mantle right now other graphic vendor might not be able to use it because of the lack of hardware needed by Mantle. even if they somehow able to make Mantle to run on their hardware it will be not as good as running it on AMD hardware.
 


In order to do a proper test, it should be done on a "full" server (64 player) and the complete round if possible to attain a better average, or say a minimum 15 minutes. Those 5 minute tests in "benchmarks" are nearly completely meaningless because 5 minutes is an insufficient amount of time for this type of test IMO (and experience in doing stats for a living). If you read some of them the testers even admit as much.

For Shanghai I use =[PT]=Total Mayhem | 24/7 Siege | Fast Spawn | High Tickets as during the evening EST it is usually full

For Metro/Op Locker I use BanZore.com 24/7 Hardcore Metro / Op Locker #1200 tickets

The procedure I use is:
1) Join server, wait until the map is fully loaded
2) In console, type perfoverlay.frametimelogenable 1
3) Play for a minimum 15 minutes, try to mix it up on map locations with as much action as possible. NO SNIPING
4) Before the round ends, type perfoverlay.frametimelogenable 0

Use BF4 Frametime Analyzer. If you see some crazy FPS highs and lows, it can skew the results. Use the smoothing option to see if it changes the averages much. If it does, the test is void right out of the gate. I make it a habit of deleting 5 (5,000 ms) seconds from the beginning and end anyway.

Cautions:
1) Do not start recording frame times until you are actually spawned on the map.
2) Do not allow the game to end without disabling frametimelog first. If the game ends or is quit before disabling, edit the data to eliminate the last 30 seconds (30,000ms) or so. Otherwise you end up with wild low/high FPS numbers.
3) Recording the game obviously skews the results. ALWAYS start recording first, then begin capturing frames and ALWAYS stop capturing before stopping recording.

I'm sure there are more ways to refine the results to test for validity, but I didn't write the app so can't say with a high degree of certainty what they'd recommend, However, I just don't think the myriads of benchmarks we see coming out of _____________ (fill in the blank) can be relied upon as much as people do.
 
First, I like 770 more, bcz I think it's more reliable in games

Even the 280x is powerful and performs very similar to 770 ..

and for both cards, they are really close and their performances in games is different

so I think they are equal ..


so ... my choice is 770

for this reason i want to know which card of 770 is the best

bcz one said ASUS doesn't OC while MSI and EVGA do ..
 


The choice is certainly yours, and in DX11 they are pretty much equal, but you're comparing apples to oranges because Mantle improves BF4 significantly.

GTX 770
https://www.youtube.com/watch?v=X3eSMT-cYTQ

GTX 780ti
https://www.youtube.com/watch?v=eNUIj3whKac

R280x (mine) on Mantle
https://www.youtube.com/watch?v=vBtXFAZXaJA

Still think they are equal? Good luck.









 
showing a 780ti dx11 running 1440p compared to a 280x mantle running 1080p, the differences are about what one would expect.... mantle can definitely narrow the gap but once a high tier cpu and above 1080p resolution is used with mantle, the differences are about as big as an overclock, but a increase none the less.
 
Status
Not open for further replies.