290 vs 390

[App4that]

So I've had the two graphics cards for awhile now. I've been looking at both very carefully. Just finished some testing and found some extremely interesting data.

 

A little backstory first. I first bought a Vapor X 290, later I bought a Nitro 390. I've run them in crossfire and alone. I've tested the temperature produced by both now and the wattage used by both.

 

All tests done at 1100MHz core and 1500MHz memory. The 290 reached 1500MHz on the memory thanks to the stability of the new Trixx software from Sapphire. 1450MHz was the previous best

 

In idle wattage usage in the case alone. Vapor-x 120w and Nitro 138w

 

At full load wattage usage in the case alone. Vapor-x 380w and Nitro 350w.

 

Idle temps. Vapor-x 47c and Nitro 42c.

 

Full load temps (synthetic benchmark) Vapor-x 76c and Nitro 62c

 

Valley FPS average in Extreme HD. Vapor-x 67.8 and Nitro 70.7

 

Now for the best part, and what fucking blows my mind. I've got over 40hrs playing Fallout 4, I know where to find dips in FPS. The Vapor-x beats the Nitro. Not by much but the GPU usage is more consistent in those spots. Where the Nitro jumps around in fps with lows in the upper 30s, the Vapor-x holds steady at 45fps. Everywhere else the Vapor-x like the Nitro pegs at the 60fps limit with max settings. Not entirely sure what to do with this information. I'm going to try Witcher 3 and keep reporting as I find new information on the two.

[Misanthrope]

This difference points to differences in the bios on the cards most likely. Also are you using the 15.11 beta driver from AMD? I tried briefly on Fallout 4 but it went from dips into 30 to solid 50 to 60 but outright freezes for half a second constantly so I uninstalled them. Not sure if this was due to some of the fiddling I did on the ini however.

[dragoon20005]

welp i say it the BIOS and the power delivery of the cards i guess

 

note that Vapor X is a custom PCB as compared to the Nitro which uses a reference PCB

[Morgan MLGman]

@App4that I told you before that those cards perform the same clock for clock m8, you didn't believe me That is why I said that with the overclock my 290X is beyond a stock 390X performance earlier

@i_build_nanosuits you might find this thread interesting

[dewman45]

It seems games just aren't optimized for the 390. I have a Gigabyte G1 Gaming 390 and the optimization isn't the best. Games will run well, but they stress the card too much. Bioshock Infinite is probably the worst. I will be getting 70-80 fps Ultra settings 1080p, and the card is the loudest I have ever heard it at 86 C. Turn vsync on, everything is fine, never dips below 60 fps at 68 C. Ark Survival runs terrible on lower settings, but if I bump them up to Epic, it runs a smooth 35-45 fps on average at 1080p with decent temps. DayZ is the worst, but I haven't even tried it with the new UI, so I won't include that. Minecraft is iffy sometimes.

[thekeemo]

welp i say it the BIOS and the power delivery of the cards i guess

 

note that Vapor X is a custom PCB as compared to the Nitro which uses a reference PCB

The nitro does not use a reference PCB

if it did EK waterblocks would be compatible

[Morgan MLGman]

It seems games just aren't optimized for the 390. I have a Gigabyte G1 Gaming 390 and the optimization isn't the best. Games will run well, but they stress the card too much. Bioshock Infinite is probably the worst. I will be getting 70-80 fps Ultra settings 1080p, and the card is the loudest I have ever heard it at 86 C. Turn vsync on, everything is fine, never dips below 60 fps at 68 C. Ark Survival runs terrible on lower settings, but if I bump them up to Epic, it runs a smooth 35-45 fps on average at 1080p with decent temps. DayZ is the worst, but I haven't even tried it with the new UI, so I won't include that. Minecraft is iffy sometimes.

Cause Gigabyte's G1 is crap. That is why your temps are high and it's loud.

[App4that]

This difference points to differences in the bios on the cards most likely. Also are you using the 15.11 beta driver from AMD? I tried briefly on Fallout 4 but it went from dips into 30 to solid 50 to 60 but outright freezes for half a second constantly so I uninstalled them. Not sure if this was due to some of the fiddling I did on the ini however.

I'm using the same latest beta driver on both, was downloaded using the 390. I was going to rinse and repeat the driver if there was a problem with the 290 but as you can see there wasn't one.

[dewman45]

Cause Gigabyte's G1 is crap. That is why your temps are high and it's loud.

Yeaaaaah... It was the cheapest one when the card came out. I could have gone with MSI, which had three month shipping. That means I may have just gotten it this month. I will take heat over time.

[Morgan MLGman]

Yeaaaaah... It was the cheapest one when the card came out. I could have gone with MSI, which had three month shipping. That means I may have just gotten it this month. I will take heat over time.

It's the cheapest one for a reason

[App4that]

OK, now this is just fucking crazy. I played Witcher 3 with just the 290. (bought Witcher 3 after I bought the 390) Hairworks, works. With the 390 Hairworks caused tentacle porn hair that shimmered like a disco ball. Using the 290 it looked like normal hair, actually really cool. (a bit of a shock going from Fallout, Witcher 3 looks 100x better :lol: ) BUT, where the 390 is in the high 50's and even pegs the fps at 60, the 290 is in the high 40's with hairworks on, off it's in the mid 50's.

 

Now what the hell do I do? Why does the 290 do a better job with gameworks games? Am I looking at keeping the 290 in the case? :huh:

[i_build_nanosuits]

@App4that I told you before that those cards perform the same clock for clock m8, you didn't believe me That is why I said that with the overclock my 290X is beyond a stock 390X performance earlier

@i_build_nanosuits you might find this thread interesting

Valley FPS average in Extreme HD. Vapor-x 67.8 and Nitro 70.7

where the 390 is in the high 50's and even pegs the fps at 60, the 290 ... it's in the mid 50's.

^^ no, they don't

[Morgan MLGman]

 

^^ no, they don't

 

 

3 FPS difference cause of faster VRAM is basically no difference. As I said, my 290X overclocks higher than stock 390X values, it's faster or at least the same, I say #worth to the change from a 970

Also "The Vapor-x (290) beats the Nitro (390). Not by much but the GPU usage is more consistent in those spots. Where the Nitro jumps around in fps with lows in the upper 30s, the Vapor-x holds steady at 45fps."

[i_build_nanosuits]

3 FPS difference cause of faster VRAM is basically no difference. As I said, my 290X overclocks higher than stock 390X values, it's faster or at least the same, I say #worth to the change from a 970

Also "The Vapor-x (290) beats the Nitro (390). Not by much but the GPU usage is more consistent in those spots. Where the Nitro jumps around in fps with lows in the upper 30s, the Vapor-x holds steady at 45fps."

you know just like i do that fallout 4 is a very wierd game with very unpredictable performance result from time to time

[Morgan MLGman]

you know just like i do that fallout 4 is a very wierd game with very unpredictable performance result from time to time

All hail Bethesda and their optimization

[i_build_nanosuits]

All hail Bethesda and their optimization

in some places in the city i've found my overclocked haswell core i7 CPU to be limiting the performance of my GTX980ti slightly on maximum settings 1440p...so yeah

[Morgan MLGman]

in some places in the city i've found my overclocked haswell core i7 CPU to be limiting the performance of my GTX980ti slightly on maximum settings 1440p...so yeah

Just a question offtopic, I see you changed your cooler, what's the reason?

[i_build_nanosuits]

Just a question offtopic, I see you changed your cooler, what's the reason?

fear the other one might leak on my 800$ graphics card soon or later...and it was starting to be low in fluid and it cost a lot to replace it...it was also louder than this one with the super nice enermax fans on it...and it really looks neat now...you should see it in real it looks amazing...you can see newer pics of my rig in a new gallery i posted in my profile but i swear it looks 10x better in real

[Morgan MLGman]

fear the other one might leak on my 800$ graphics card soon or later...and it was starting to be low in fluid and it cost a lot to replace it...it was also louder than this one with the super nice enermax fans on it...and it really looks neat now...you should see it in real it looks amazing...you can see newer pics of my rig in a new gallery i posted in my profile but i swear it looks 10x better in real

Yeah, I saw you updated your avatar so I went into your gallery and checked it out, looks dope ^^ The cooler isn't as efficient as Triton though, is it? I've got a beefier one, it's a twin-tower style with also 2x120mm fan and I love it, it has worse quality fans but probably performs a bit better in terms of cooling cause of nearly twice as big rad

[App4that]

 

^^ no, they don't

 

The thing you can't post as a number is the look of the hair, even with the fps loss the 290 does a better job with the game. And the fps difference doesn't translate to anything you'd notice, I almost don't trust the fps meter in Steam it ran that smooth. I hate to admit it, I mean REALLY hate to admit it since I own both and could have got a 4790k for the price I paid for the 390, but Mr. MLGman might be right.

[i_build_nanosuits]

Yeah, I saw you updated your avatar so I went into your gallery and checked it out, looks dope ^^ The cooler isn't as efficient as Triton though, is it? I've got a beefier one, it's a twin-tower style with also 2x120mm fan and I love it, it has worse quality fans but probably performs a bit better in terms of cooling cause of nearly twice as big rad

yes the triton had better temps, prolonged full load temps went from 66c to 72c and gaming temps went from 55c to 62c...not big deal to be honest my overclock is very efficient i achieve 4.2ghz with only 1.18v on the core that's why it won't change that much...i need like 1.3v for 4.3ghz stable so really not worth it at that point...my 4770k is on the lower side for overclocking capabilities but it run cool and fast at those settings so i'm happy with it.

And yes a bigger rad will have better temps...i don't like watercoolers i'm done with them...

[dragoon20005]

The nitro does not use a reference PCB

if it did EK waterblocks would be compatible

either my eyes are playing trick on me or the Nitro PCB looks like this

 

 

 

vs the Vapor X PCB

 

[thekeemo]

either my eyes are playing trick on me or the Nitro PCB looks like this

 

vs the Vapor X PCB

 

Im not saying that they are the same. Just saying that neither is reference.

[App4that]

either my eyes are playing trick on me or the Nitro PCB looks like this

vs the Vapor X PCB

Damn! Naked Vaporx sexiness! I dropped the temps of the Vaporx by dropping it to the second PCIe which dropped it bellow the fan attached to the HD cage so that it blew across the backplate. This also placed it closer to the intake fan in the bottom of the case. No change in performance, just a drop is temps. It's at 70c running Fallout 4. Still way up from the Nitro.

[xAcid9]

either my eyes are playing trick on me or the Nitro PCB looks like this

 

 

 

vs the Vapor X PCB

 

 

Reference look like this.

 

 

XFX 390x

 

Sapphire 290x Tri-X