• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Poll: Fury X Owners. Would you buy the Fury X again if you could do it all over again?

With hindsight, would you buy the Fury / Fury X again?

  • Yes, I would.

    Votes: 31 36.0%
  • No, I wouldn't.

    Votes: 55 64.0%

  • Total voters
    86
That has somewhat changed. With the newer drivers at 1440P it matches the stock 980 Ti on average.

9908c0_81b8a4ecf4c546eaa427e1ea41289296.png

That's good news all around :)

The problem comes when comparing the FX to 'out of the box' overclocked Ti cards, they tend to bury the FX by 25-30% sometimes :(


This very recent review gives an idea;

http://hexus.net/tech/reviews/graphics/89717-evga-geforce-gtx-980-ti-hybrid-gaming/
 
[/spoiler]

That's good news all around :)

The problem comes when comparing the FX to 'out of the box' overclocked Ti cards, they tend to bury the FX by 25-30% sometimes :(


This very recent review gives an idea;

http://hexus.net/tech/reviews/graphics/89717-evga-geforce-gtx-980-ti-hybrid-gaming/

Yes. That's the issue. They were comparing it with the "stock" 980 Ti. I mean I got mines less than $60 - $70 less than the 980 Ti stock.

I can see why many would get an overclocked out of the box 980 Ti. Heck I even brought a XFX 7600 GT XXX edition which was an overclocked 7600 GT. A lot of people are doing the same with slightly overclocked 980 Ti and paying a little more.

I think to counter this problem AMD's Board partners need to release an OC'ed version of the Fury X like MSI Lighting Fury X similar to what their partners did with the R9 290X. R9 290X's sold very well.

I wonder why they are not doing so? There were manufacturing issues 3+ month's ago according to AMD's CEO with the Fury X. Understandable I suppose as they would have to go through an extra layers where they need to get the pump + closed loop liquid cooler. But that has seen been cleared up.

Anyone want to venture a guess as to why AMD's board partners are not releasing OC'ed Fury X? I think might help with the sales.
 
Yes. That's the issue. They were comparing it with the "stock" 980 Ti. I mean I got mines less than $60 - $70 less than the 980 Ti stock.

I know it is relevant to you but "your" purchase cost is not remotely a factor when comparing new vs new prices. An R9 Fury goes for around the same price as a 980Ti and having owned both and considering the extra 2GB VRAM and the superior overclocking headroom the 980Ti is a better all round GPU. The only reason to get a Fury X over a 980Ti is if you are a fanboy, don;t overclock (where it's competitive) or are locked in to Freesync. I fall in to the latter category and as such have settled on a Fury (non X) because I simply don't trust I would get a Fury X without pump whine.

I can see why many would get an overclocked out of the box 980 Ti. Heck I even brought a XFX 7600 GT XXX edition which was an overclocked 7600 GT. A lot of people are doing the same with slightly overclocked 980 Ti and paying a little more.

I think to counter this problem AMD's Board partners need to release an OC'ed version of the Fury X like MSI Lighting Fury X similar to what their partners did with the R9 290X. R9 290X's sold very well.

I wonder why they are not doing so? There were manufacturing issues 3+ month's ago according to AMD's CEO with the Fury X. Understandable I suppose as they would have to go through an extra layers where they need to get the pump + closed loop liquid cooler. But that has seen been cleared up.Anyone want to venture a guess as to why AMD's board partners are not releasing OC'ed Fury X? I think might help with the sales.

You have essentially asked the same question 3x times there. :)

The answer is Fury X has been locked down by AMD, they will not allow custom versions. Only Fury non X can be customised by AIBs and there are overclocked versions with improved/custom PCBs.
 
Last edited:
I know it is relevant to you but "your" purchase cost is not remotely a factor when comparing new vs new prices. An R9 Fury goes for around the same price as a 980Ti and having owned both and considering the extra 2GB VRAM and the superior overclocking headroom the 980Ti is a better all round GPU. The only reason to get a Fury X over a 980Ti is if you are a fanboy, don;t overclock (where it's competitive) or are locked in to Freesync. I fall in to the latter category and as such have settled on a Fury (non X) because I simply don't trust I would get a Fury X without pump whine.

a) 2GB VRAM doesn't mean a lot, considering even at demanding 4K, the FuryX kicks the 980Ti because it has much much faster memory.
And weren't you the other day saying that the 3.5GB GTX970 is fine against a 8GB 390 and the extra ram not needed?

b) A FuryX goes for £389 brand new, as R9 Nano. That is £110 LESS than the cheapest 980Ti, and with £70 waterblock you have a small size custom cooled Fury X at £460. Or a normal AIO FuryX at £500. The Watercooled versions of the 980Ti start at £600 alone.....
 
None of it negates the fact that with the only change being I'm now on the latest Crimson drivers my Fury Tri-x scores roughly the same with an 1130 overclock as it did on it's 1040 box clock 4 months ago in Firestrike extreme.
So where are these driver improvements, It seems like a step backwards to me. Admittedly it's not a game but without voltage control I was able to get an 1100 stable clock for Firestrike but only 1070 for 100% stability in games (Some games crashed at 1080mhz).

At the end of the day the fact that I run my PC on stock clocks is just as well, It would have been nice to have been able to put some decent scores up on the forum bench threads though. I've done a couple just to see 290x's above them due to the fact they clock better.
 
a) 2GB VRAM doesn't mean a lot, considering even at demanding 4K, the FuryX kicks the 980Ti because it has much much faster memory.
And weren't you the other day saying that the 3.5GB GTX970 is fine against a 8GB 390 and the extra ram not needed?

The reason Fury X trades blows with 980Ti at 4K is because you absolutely must compromise graphical settings on both to get playable FPS, so the VRAM limit is not getting reached on single GPU. The extra 2GB VRAM comes into use if you plan to go multi GPU and as such for most people at this level it is a factor.

b) A FuryX goes for £389 brand new, as R9 Nano. That is £110 LESS than the cheapest 980Ti, and with £70 waterblock you have a small size custom cooled Fury X at £460. Or a normal AIO FuryX at £500. The Watercooled versions of the 980Ti start at £600 alone.....

The Fury X does not cost £389, that's a Nano and in stock form it is slower than an Fury non-X due to TDP constraints and your £70 cost for the waterblock does not include the considerable cost of the pump, fluid, radiator and tubing etc. If you want a stock Fury X it will cost you ~£500 as will a good custom cooled 980Ti. The fact it has AIO water cooling as standard is not necessarily a bonus to many people as all they can see is a GPU that has less overclocking potential compared to 980Ti. OC vs OC it is around 10-15% slower than 980Ti (at 4K where I tested a Fury, Fury X and 980Ti).
 
Last edited:
None of it negates the fact that with the only change being I'm now on the latest Crimson drivers my Fury Tri-x scores roughly the same with an 1130 overclock as it did on it's 1040 box clock 4 months ago in Firestrike extreme.
So where are these driver improvements, It seems like a step backwards to me. Admittedly it's not a game but without voltage control I was able to get an 1100 stable clock for Firestrike but only 1070 for 100% stability in games (Some games crashed at 1080mhz).

At the end of the day the fact that I run my PC on stock clocks is just as well, It would have been nice to have been able to put some decent scores up on the forum bench threads though. I've done a couple just to see 290x's above them due to the fact they clock better.

The problem is you are setting yourself up for a disappointment if you are referring to synthetic benchmarks. In Witcher 3 I am seeing 10-13% extra performance at 4K from my overclocked Fury pro OC to 1150/550 which is right in line with what I would expect. By comparison my 290X Lightning at 1200 core, or my 980 at 1350 core got anywhere near playable at similar settings at 4K.
 
The problem is you are setting yourself up for a disappointment if you are referring to synthetic benchmarks. In Witcher 3 I am seeing 10-13% extra performance at 4K from my overclocked Fury pro OC to 1150/550 which is right in line with what I would expect. By comparison my 290X Lightning at 1200 core, or my 980 at 1350 core got anywhere near playable at similar settings at 4K.

That's another issue cause I'm not at 4k and have no intention of going 4k because the truth is the current crop of cards struggle at 4k. I tried quite a few games at 4k via vsr and it's dire so I run at 1440p via vsr.

I don't want my fps sitting in the 30 zone and I like my games to look as visually appealing as possible.
The fact you overclock the memory is great but like a lot of people I can't even get it too 501 without blue screening (I haven't tried with AB voltage though, just Trixx, I'll have to give that a go later).
I'm not sure how comparing how my performance has changed using the same program is setting myself up for anything. Today my Firestrike extreme result matches what it did 4 months ago and that was with my Fury at it's stock clock 4 months ago, Today it's got an 1130 overclock. It's nothing to do with gaming, It's an observation regarding how performance has changed over the last 4 months with that program.
 
Last edited:
That's another issue cause I'm not at 4k and have no intention of going 4k because the truth is the current crop of cards struggle at 4k. I tried quite a few games at 4k via vsr and it's dire so I run at 1440p via vsr.

I don't want my fps sitting in the 30 zone and I like my games to look as visually appealing as possible.
The fact you overclock the memory is great but like a lot of people I can't even get it too 501 without blue screening (I haven't tried with AB voltage though, just Trixx, I'll have to give that a go later).
I'm not sure how comparing how my performance has changed using the same program is setting myself up for anything. Today my Firestrike extreme result matches what it did 4 months ago and that was with my Fury at it's stock clock 4 months ago, Today it's got an 1130 overclock. It's nothing to do with gaming, It's an observation regarding how performance has changed over the last 4 months with that program.

I absolutely agree about 4K being too much for the current crop of high end GPUs (Fury X, Titan X and 980Ti) for non adaptive sync monitors. I have a 4K adaptive sync monitor and 35 - 50 FPS on it is superior to 45-60 FPS without it. It has made the drop from 980Ti to Fury pro bearable and in fact is superior as a gaming experience. At the time there was no 4K 32" g-sync monitors for me to use with the 980Ti I had so my only option was Fury and Freesync.
 
That's another issue cause I'm not at 4k and have no intention of going 4k because the truth is the current crop of cards struggle at 4k. I tried quite a few games at 4k via vsr and it's dire so I run at 1440p via vsr.

I don't want my fps sitting in the 30 zone and I like my games to look as visually appealing as possible.
The fact you overclock the memory is great but like a lot of people I can't even get it too 501 without blue screening (I haven't tried with AB voltage though, just Trixx, I'll have to give that a go later).
I'm not sure how comparing how my performance has changed using the same program is setting myself up for anything. Today my Firestrike extreme result matches what it did 4 months ago and that was with my Fury at it's stock clock 4 months ago, Today it's got an 1130 overclock. It's nothing to do with gaming, It's an observation regarding how performance has changed over the last 4 months with that program.

I take it you increased the power limit?
It sounds like you're overclock is throttling, as it's being consumed by the power limit as you should see about 6-7% performance increase from your clockspeed.
Also are you looking at the graphics score and not the p score.
 
Last edited:
Just over 100 votes now, I was actually fairly interested to see what peoples thoughts were, but what's blew me over is the fact over 100 of these sold !
 
I know it is relevant to you but "your" purchase cost is not remotely a factor when comparing new vs new prices. An R9 Fury goes for around the same price as a 980Ti and having owned both and considering the extra 2GB VRAM and the superior overclocking headroom the 980Ti is a better all round GPU. The only reason to get a Fury X over a 980Ti is if you are a fanboy, don;t overclock (where it's competitive) or are locked in to Freesync. I fall in to the latter category and as such have settled on a Fury (non X) because I simply don't trust I would get a Fury X without pump whine.
There can be other reasons other than being a fanboy. I'll give you 3 to prove it.

1) your perspective and expected use. For me I'm okay to give 10% or so performance difference (don't really view 3fps ontop of 30, 2fps ontop of 20, 4 fps ontop of 40 as a major difference for example) if it means I can back the underdog / other stuff I like. I've always backed the underdog as I prefer to give competition and not watch the industry become a monopoly (boo apple) but even then my perspective isn't just support the poor guy for the sake of it but if it's worth supporting and to be fair performance is always give and take depending on particular games so sometimes (like with star wars battlefront) you are better picking the card of your choice. AMD has some benefits like freesynch, better image quality and I simply like there design ethic (open source and clean along with generally more affordable gpu's like 390 being better than 970).

2) if you don't want little nibbles or care about supporting the underdog then how about actual performance to price or rather experience. In my eyes AMD are better than Nvidia price wise on everything from the 390 downward but if we're specifically looking at experience from fury / 980ti then as you pointed out I CAN get a better experience overall if I buy fury anyway thanks to freesynch.

https://www.overclockers.co.uk/lg-2...g-widescreen-led-monitor-black-mo-138-lg.html

https://www.overclockers.co.uk/iiya...-gaming-widescreen-led-monitor-mo-127-iy.html

anyone care to find a gsynch monitor of similar price that can compare to these? In my eyes short of having more money than is needed (and also spending more than is sensible thereby blowing price to performance so lessening the value) you can't really get a gsynch monitor that does compare to that quality. You'll have to settle for 1440p or settle for none ips or whatnot. My overall experience / value is better with the AMD option still as exampled there so we have experience and price to performance as you get 4k out of the deal at a lower price. You can probably got freesynch with Nvidia soon anyway but I forget if the HDMI cable is enough to actually get 4k anyway.

3) if price to performance, experience and supporting the underdog / competition / open standards are all not your thing then the final point is that AMD simply are an alternative. Not everyone has a great experience with AMD but equally not everyone has a great experience with Nvidia either. It's a minor point but to be fair the argument on performance for price (with the experience of 4k and freesynch not overcharging you), some games performing better on AMD as it can be hit and miss so you should always look at the cards that support the games you intend to play and the notion that overclocking providing maybe a 10% difference really isn't that big a loss then I feel you really can benefit from the AMD option still even when focusing on performance. Should you not be looking to set up a 4 way SLI titan build with a gsynch monitor then there's no real reason the AMD option can't work but to be fair I acknowledge it helps more if you're aiming for a freesynch monitor and not the kind of person spending 5k on your next rig just to impress.

just adding my 2 cents to this argument on AMD being behind but all in all the 980ti is the better card, thankfully AMD has made it not all about the card so we can get a better experience for less money regardless. I'm not speaking from experience as I don't have a fury but to me it makes sense that freesynch and gsynch have changed the game and people SHOULD be looking at those monitors if they are going to be spending 400 or 500 on a gpu.
 
Last edited:
I know it is relevant to you but "your" purchase cost is not remotely a factor when comparing new vs new prices. An R9 Fury goes for around the same price as a 980Ti and having owned both and considering the extra 2GB VRAM and the superior overclocking headroom the 980Ti is a better all round GPU. The only reason to get a Fury X over a 980Ti is if you are a fanboy, don;t overclock (where it's competitive) or are locked in to Freesync. I fall in to the latter category and as such have settled on a Fury (non X) because I simply don't trust I would get a Fury X without pump whine.

Well the price of the Fury X/Fury varies from country to country. I got it cheaper $60 - $70 than the 980 Ti as the primary reason. I think it may have been a difficult choice if it was priced the same. Funny thing is when they dropped the price of the Fury X like the Sapphire Fury X to around $570 on newegg it all got sold out quickly during the holidays. Which goes to show that if they had launched it at that price I think it would have generated much greater "mind share" and enthusiasm and thus more sales.

With respect to overclocking, I don't really overclock much. As the Fury X is more than fast enough "currently" as such my out of the box experience would be slightly better than a stock 980 Ti which would have cost me $50 - $70 more. I only overclock when I need to. I overclocked my HD 6970 only when I needed to get the extra performance. I know that I may be the exception but I know others who have brought stock 980 Ti's and haven't overclocked.

With respect to the extra 2GB. I can definitely see that as a reason as I also know another person who brought the 980 Ti and he sighted the extra 2GB as the reason as 4GB was pushing it at 4K in GTA V as he put it. But let's face it, the extra 2GB I don't think will matter much. Going forward, if you wanted a GPU that will last you at least 2+ years that will allow you to play games maxed out at 4K with at least FXAA you realistically need 8GB+ plus more horsepower than the 980 Ti/Fury X.

My other main reason was similar to your last reason and that is FreeSync. I don't want to be locked down to one vendor and would like to have an open and standard version of Variable Refresh rate technology. I am thinking of getting a FreeSync monitor and not only that it is cheaper than G-Sync by almost $150 - $200. Not worth the extra cost for essentially the same technology.
 
Last edited:
Anyone what to take a stab at this question?

The Fury X has 176 GB/S more bandwidth than the 980 Ti but it doesn't pull away from the 980 Ti buy a significant margin. I found that a bit shocking to be honest.

I was thinking that with Fury X's higher bandwidth memory it would have easily beat the 980 Ti.

I was thinking this would be a repeat of the HD 4870 when it launched with the World's First GDDR5 RAM. It beat the GTX 260 on average which had a memory bandwidth of around 111 Gb/s vs 115GB/s of the HD 4870 and it even beat the GTX 280 in some games while it cost's $100 less than the GTX 260 and almost half the cost of a GTX 280 when it launched.

I was totally floored by the price/performance of the HD 4870 and brought my first ATI card!

Given the specs of the Fury X with it's insane 176GB/s bandwidth advantage and 4096 Bit bus, shouldn't it have demolished the 980 Ti. I know there are architectural differences between the two GPU's and memory bandwidth shouldn't be the only factor in performance but still. On paper the hardware of the Fury X looks vastly superior but it doesn't blow away the 980 Ti in terms of performance.

You guys think it's a driver issue or something?
 
I agree, having the Nano aircooled still, no overclocks etc all default, and feels powerfull.
And the EK prefilled block is here, but waiting OCUK to have Predator 360 rev 1.1 at stock
 
Anyone what to take a stab at this question?

The Fury X has 176 GB/S more bandwidth than the 980 Ti but it doesn't pull away from the 980 Ti buy a significant margin. I found that a bit shocking to be honest.

I was thinking that with Fury X's higher bandwidth memory it would have easily beat the 980 Ti.

I was thinking this would be a repeat of the HD 4870 when it launched with the World's First GDDR5 RAM. It beat the GTX 260 on average which had a memory bandwidth of around 111 Gb/s vs 115GB/s of the HD 4870 and it even beat the GTX 280 in some games while it cost's $100 less than the GTX 260 and almost half the cost of a GTX 280 when it launched.

I was totally floored by the price/performance of the HD 4870 and brought my first ATI card!

Given the specs of the Fury X with it's insane 176GB/s bandwidth advantage and 4096 Bit bus, shouldn't it have demolished the 980 Ti. I know there are architectural differences between the two GPU's and memory bandwidth shouldn't be the only factor in performance but still. On paper the hardware of the Fury X looks vastly superior but it doesn't blow away the 980 Ti in terms of performance.

You guys think it's a driver issue or something?

In the memory stakes the GM 200 cards totally thrash the fury X. It was a huge mistake for AMD to use HBM1 @500mhz.

Mhz is more important than bandwidth most of the time. Fortunately HBM2 will address the Mhz problem.

The Fury X is a good card but why are you pushing it so hard, what is your angle ?
 
Back
Top Bottom