Possible Radeon 390X / 390 and 380X Spec / Benchmark (do not hotlink images!!!!!!)

[Mal X]

if you're thinking about a 390X Rebrand............aren't you better off with the MSI Lightning 290X ??????

whatever the case it's pretty lame for the same card and a year later, because you'd expect it to be 6% faster, not the same !!!

 

[N19h7m4r3]

I like how the folks harping about Furmark ignore the rest of the tests that were run.

Wasn't long ago Anandtech and others were still running Furmark as well. I'll never forget endless harping people gave the likes of the Mac Pro because Furmark made it run hot and noisy, and it only did it there.

Can't say I care for it, but the chap was nice enough to buy the card and run a bunch of tests for us. Especially since reviewers didn't get sent any cards at all.

 

[ICDP]

That one was $369.99 + tax at best buy for the 390.

Straight conversation is about £250 but we all know it will be nearer £300

Best Buy typically add around $70 - $90 to the same price you can a card from etailers such as Newegg.

http://www.newegg.com/Product/Product.aspx?Item=N82E16814150696

http://www.bestbuy.com/site/xfx-rad...1310401532.p?id=mp1310401532&skuId=1310401532

So about £193 + 20% Vat = ~£230, or a similar price to the current 4GB models of 290X.

 

[tommybhoy]

That's a 390 and not an x against an out of the box overclocked 970. It is scoring a decent bit better than my 1040/1350 290 in the overall/graphics score. It looks like it will be faster than a stock 970 or there about which we all know a standard 290 isn't. I am still not convinced it's a straight 290 after seeing those results.

290X v stock 970 comparison:

http://www.3dmark.com/3dm/6994971?

http://www.3dmark.com/3dm/6995377?

 

[StrongSad]

Will the 390 be around £230-£240 ?

It must be around that price IMO or it qill seem little value for money.

There is a $20 difference between the 290 and the 970
a $45 difference between a 970 and 390
and $65 between the 290 and 390. (the difference is between the AMD is on XFX models 290 - 390)

 

[Nutella33]

Can't say I care for it, but the chap was nice enough to buy the card and run a bunch of tests for us. Especially since reviewers didn't get sent any cards at all.

No, just he hasn't got a card. That doesn't mean all reviewers don't have one.

He also says he doesnt care about unreleased products and doesn't follow the news and then says companies should hand out unreleased products whilst running a site that has "Tech new and reviews" sections.

Shipment of 24 Fury X's spotted on the 8th of june btw, Might be connected with E3 and handing out to reviewers there.

 

[TheRealDeal]

290X v stock 970 comparison:

http://www.3dmark.com/3dm/6994971?

http://www.3dmark.com/3dm/6995377?

That 290x is running 1050/1350 which is not stock. Stock is 1000/1250 or were you getting the 290x to clocks where it was a match for the 970.

 

[TonyTurbo78]

think I'll wait for the rebranded 490X with 1100MHz core and 6500MHz clocked VRAM

 

[Boomstick777]

1 Day to go !

 

[RanxZy]

That 290x is running 1050/1350 which is not stock. Stock is 1000/1250 or were you getting the 290x to clocks where it was a match for the 970.

Regardless, in game our eyes would notice very little difference between a hundred firstrike points or 50MHz difference in clock speed.

 

[ubersonic]

Regardless, in game our eyes would notice very little difference between a hundred firstrike points or 50MHz difference in clock speed.

I noticed the 50MHz difference between the Riva 128 and the Riva TNT2 Ultra

 

[Mayh3M]

think I'll wait for the rebranded 490X with 1100MHz core and 6500MHz clocked VRAM

 

[tommybhoy]

That 290x is running 1050/1350 which is not stock. Stock is 1000/1250 or were you getting the 290x to clocks where it was a match for the 970.

Never said the 290X was stock, just the 970, it had a slightly higher core clock(which is a better core comparison to the 390X reported clocks) to match my 290X DCuII clocks in CrossFire.

 

[TheRealDeal]

Regardless, in game our eyes would notice very little difference between a hundred firstrike points or 50MHz difference in clock speed.

Most likely not but in reviews most people will be looking to whats faster whether that's 100 points or 1, faster is better in most people's eyes.

 

[Ayahuasca]

Most likely not but in reviews most people will be looking to whats faster whether that's 100 points or 1 faster is better in most people's eyes.

Agreed, people see higher numbers and it makes them buy. Doesn't matter if it looks no different in game as long as you can post a higher score

 

[TheRealDeal]

Never said the 290X was stock, just the 970, it had a slightly higher core clock(which is a better core comparison to the 390X reported clocks) to match my 290X DCuII clocks in CrossFire.

If the 390 is nearly matching a 970 at 1000/1500 then the 390x at 1050/1500 with more cores is going to be a good bit faster. This is going by that guys results. His 390 score is making me think that there have been changes to the core even if they are minor.

 

[LoadsaMoney]

IF 4GBs only good enough for 1080p, then Furys going to be no good for FreeSync, as all the bloody monitors are 2560/1080,1440p and up, ouch!

 

[Rroff]

IF 4GBs only good enough for 1080p, then Furys going to be no good for FreeSync, as all the bloody monitors are 2560/1080,1440p and up, ouch!

4GB is fine for 2560x1440 - dunno for how long as with DX12, etc. that gets a bit complicated - ultimately though even with tiled/streaming resources, etc. I can't see developers just leaving VRAM sitting there unused when there are plenty of other things they could make it do none the least increase the detail and complexity of animations, etc.

UHD/multi high res panels is another matter though...

 

[Ayahuasca]

I'm a HBM believer, AMD are not stupid enough to gimp their own top card with 4GB if it doesn't work well at 4k. It was probably the first thing they tested.

 

[flopper]

I'm a HBM believer, AMD are not stupid enough to gimp their own top card with 4GB if it doesn't work well at 4k. It was probably the first thing they tested.

You see the issue with Titanx and witcher 3 for example as the card cant sustain 60 fps maxed out nor can 980ti at 1080p!!!!!! why have 6 or 12gb when the card cant power up enough performance?

While DX12 changes this equation a bit Fury has enough ram with 4gb as nothing else indicate differently.