Try harder to read the charts you actually link, because it actually shows the complete opposite of what you thought.
According to Anandtech the same is true:
Here's the comparison that matters right now when they're equivalently priced at $650 each (980Ti vs Fury X):
http://www.anandtech.com/bench/product/1496?vs=1513
And as someone who's bought AMD for close to a decade now, because they've been rock-solid GPUs that performed well and had a solid bang-for-the-buck, I think the 980Ti is the better choice. You get 2GB more VRAM, better performance in 99% of the stuff tested, quieter, cooler, and you get a nice air-cooled card rather than some hydro contraption. Oh yeah, and HDMI 2.0, right?
This is so old and tired, but it always gets me lol-ing even after all this time:
https://www.youtube.com/watch?v=xhVo7yPjQvE
Sorry AMD, we still love you though!
That line at the end "AMD let me down again but I still don't mind saving money buying their alternatives" is so true. I have had several AMD cards and one AMD CPU all but one of them were either good enough for the cost or an outright disappointment but I still keep buying them based on price.
AMD is in a tough spot I don't see myself spending a premium on any of their products now.
TBH AMD could probably release Fury X at $1000 and milk the hard core fanboys/filthy rich just like the Titan X and Titan did, then lower the price to something reasonable once they get sufficient volume to actually worry about not selling out.
No, the benchmarks don't support that at all. That's what made the Titan X screw-job so brilliant, they released it in a vacuum...no other GPUs anywhere near it at the time, which made it look like a value.
Yet here we are just 3 months later and every Titan X owner is thinking "why the F did I buy this..."
I had one and luckily sold it to my cousin for what I paid for it, as it was a great overclocker. Others aren't so lucky, I see that this week's Titan X sales on ebay have closed/finalized at an average of $910, which is a large loss considering ebay then takes 10% of that figure as their own fee.
Fury X at $1000 wouldn't have sold a single card, ever. But...if it had released a month before Titan X? Oh definitely, they could have moved thousands.
Timing is everything, which AMD never, ever, seems to understand.
It has literally nothing to do with benchmarks.
The Fury X sold out in 5 minutes losing to the 980 Ti. Fanboys gonna fanboy. They already buy FX cpus which are undeniably far worse vs the competition than Fury X.
I guarantee they would have sold out at $1000 given how small the supply was. There's a rumor Newegg literally had 100 total...
Nope, dead wrong.
A thousand bucks is a thousand bucks, and yes you'd have a few, very few, that would fanboy it up and drop a $1k on a clearly inferior value ($1k FX versus $649 980 Ti), but nothing close to Titan X (again, vacuum launch, fastest GPU in the world by a wide margin back then).
I disagree completely. Fanboys/the very rich bought Titan X @ $1000 knowing a cheaper version with the same performance would come it in 2-3 months.
It would have sold out at $1000. I dont care one tiny bit that you disagree. I think you're wrong. You think I'm wrong. Nothing will convince us otherwise. So its pointless to continue this discussion.
See post #136
If chalking this up as a win for you will make you feel good man, do whatever you like
How do I do a thumbs up emotican haha.
Either way your thoughts on timing are correct though as a base. By gambling on needless (does not affect fps at all) HBM memory, AMD was late with this GPU to market. And that lateness has now done the following:
1. Killed profit margin (a Fury X launched when Titan X launched could have sold for $799 and looked like a tremendous value).
2. Heaped bad PR on a company on the brink of hostile acquisition, stock now down -10% versus prior to FX launch.
3. Disappointed their strongest support base (fans) by overhyping the impact of HBM as some sort of competitive advantage prior to launch, when the benchmarks now clearly show memory bandwidth has little impact on actual game performance (GDDR5 good enough in other words).
4. Launched a 4GB product into a competitor's 6GB product, which as silly as it sounds to informed people like ourselves, will dramatically cause lower sales because the majority of even light enthusiast customers will perceive they're paying the same price for a lower-spec card.
Basically in doing this little list, HBM killed the Fury X. That GPU was strong enough that they didn't need to screw around with advanced (and difficult to produce) newfangled memory. It made them late to market, they couldn't get a 6GB/8GB product out the door, and now the backlash is heavy.
If they had done a $599 GDDR FX, and a $499 GDDR Air Fury, both with 8GB...and launched in March because no HDM delay...wow. That would have crushed it I think, just blown the doors off at that point.
How do I do a thumbs up emotican haha.
Either way your thoughts on timing are correct though as a base. By gambling on needless (does not affect fps at all) HBM memory, AMD was late with this GPU to market. And that lateness has now done the following:
1. Killed profit margin (a Fury X launched when Titan X launched could have sold for $799 and looked like a tremendous value).
2. Heaped bad PR on a company on the brink of hostile acquisition, stock now down -10% versus prior to FX launch.
3. Disappointed their strongest support base (fans) by overhyping the impact of HBM as some sort of competitive advantage prior to launch, when the benchmarks now clearly show memory bandwidth has little impact on actual game performance (GDDR5 good enough in other words).
4. Launched a 4GB product into a competitor's 6GB product, which as silly as it sounds to informed people like ourselves, will dramatically cause lower sales because the majority of even light enthusiast customers will perceive they're paying the same price for a lower-spec card.
Basically in doing this little list, HBM killed the Fury X. That GPU was strong enough that they didn't need to screw around with advanced (and difficult to produce) newfangled memory. It made them late to market, they couldn't get a 6GB/8GB product out the door, and now the backlash is heavy.
If they had done a $599 GDDR FX, and a $499 GDDR Air Fury, both with 8GB...and launched in March because no HDM delay...wow. That would have crushed it I think, just blown the doors off at that point.
Trying to look at this from a different perspective:
Possibly knowing they'd be late to the market with a lineup consisting mostly of refreshes maybe they used this as a cleaning of the pipes for HBM. They are likely switching nodes for their next lineup, might as well work out HBM now. Maybe they know HBM will be in short supply and don't really plan on selling a lot of Fury products anyway. In the end this is still a more competitive product against Maxwell than what they had, maybe it's only intended as a stop gap until the next full lineup. They did made efficiency strides and confronted both the heat and noise concerns. Which also contributed along with HBM to significantly reduce the ever more monstrously growing size of the modern day video card which is a problem considering the growing trend of trying to make the overall size of the typical PC smaller (mini-ITX). *shrug*
Trying to look at this from a different perspective:
Possibly knowing they'd be late to the market with a lineup consisting mostly of refreshes maybe they used this as a cleaning of the pipes for HBM. They are likely switching nodes for their next lineup, might as well work out HBM now. Maybe they know HBM will be in short supply and don't really plan on selling a lot of Fury products anyway. In the end this is still a more competitive product against Maxwell than what they had, maybe it's only intended as a stop gap until the next full lineup. They did made efficiency strides and confronted both the heat and noise concerns. Which also contributed along with HBM to significantly reduce the ever more monstrously growing size of the modern day video card which is a problem considering the growing trend of trying to make the overall size of the typical PC smaller (mini-ITX). *shrug*
This is a wild guess based on no truth whatsoever: but the price gap between the Titan and the 980 Ti makes my wonder if somebody at Nvidia knew what AMD had in store and cut the price on the Ti just because "they can." Sure it hurts the sales of the cash cow Titan, but what they are doing to AMD right now will be even more profitable.
They seemingly needed the HBM for the attached power saving, giving them more scope for a faster GPU. What would be the point if it was 10/20 per cent slower?
The obvious, non trivial drawbacks are rather unfortunate. Makes the timing seem odd in some ways, but I guess they felt they couldn't really wait longer.