I have to say I was a bit shocked by some of the responses to this post.Seen on HardOCP: http://www.hardocp.com/article/2016/05/09/nvidia_founders_edition_cards_yea_or_nay
"I have literally been told for years, that NVIDIA either breaks even or loses money on NVTTM/reference cards sold into the market."
Well look at that. So this explains the $100 premium.
Ripleys effect.
There seems to be some confusion on the vapor chamber on Nvidia reference cards. Nvidia has been using vapor chambers on 250W top end reference cards since 580, but on 680,770,970,980, though those reference coolers looked just like the Titan/780/780ti/980ti they did not have a vapor chamber.
So the midrange refernce 1080 having a vapor chamber is new for that 180w power class.
So what? Do you really think it needs a $100 premium otherwise Nvidia will lose money? Talk about a captured audience.There seems to be some confusion on the vapor chamber on Nvidia reference cards. Nvidia has been using vapor chambers on 250W top end reference cards since 580, but on 680,770,970,980, though those reference coolers looked just like the Titan/780/780ti/980ti they did not have a vapor chamber.
So the midrange refernce 1080 having a vapor chamber is new for that 180w power class.
Well, GP104 is ~55% of GM200's size, and the 1080 is supposedly 20% faster than a Titan X. The proposed 232mm² die is almost the exact same ratio to Hawaii, so an uplift over Hawaii isn't out of the question.
The other part of P10's market position is more questionable. Ignoring the ludicrous pricing of the Titan series, the 1080 is launching at 108% of the launch price of the 980Ti. Launch price of the 390X was $430, and launch price of the 290X was $550. I don't think we'll see P10 launch for more than the 390X, but I also don't think we'll see it launch for 70% of the 390X's launch price. Not if it does give +20% on a 390X's performance.
So what? Do you really think it needs a $100 premium otherwise Nvidia will lose money? Talk about a captured audience.
Prices are still insane and being defended.
They've been standard for quite some time... That's what has me laughing.
Heck my bitcoin mining rig used 3 Sapphire VaporX 7970s.
I think it's best to wait for the reviews before we go on any further. Maybe there's a logical reason for the drastic price hike.
my question is why not wait for benchmarks? and why is there a time window for your build? especially when you don't have failing parts that forces you to buy?I think NV knows the market right now. After the whole 970 fiasco, I was really wanting to go AMD this round, but not sure I will have a high-performance option. My build is happening at the end of this month, and timing will be perfect for the 1080/1070. Not sure P10 will have enough performance for me (was on the fence for the 980Ti already) and the 1080 will probably get my $$$. Hopefully will get the chance to reassess the landscape at the end of the year or early next year and see if there are other options.
Jesus Christ the 980Ti Extreme Gaming is 15 times faster than my current graphics card.
Where'd you see him owning, or mentioning, that 980 :hmm:I had no idea there was that much of a performance gap between a normal 980 and some of the aftermarket 980 Ti's...that's ridiculous.
My apologies.At no point did I comment on the $100 price increase. Just setting the record straight on what seemed to be in dispute about the vapor chamber.
my question is why not wait for benchmarks? and why is there a time window for your build? especially when you don't have failing parts that forces you to buy?
You are in a good spot.I will def wait for benchmarks, but it appears unlikely p10 will be delivering above 980Ti performance (but definitely will wait and see).
My situation will be tough. Will have a VR headset sitting on my desk begging for a GPU. That is tough on my patience.
Yeah not 980.I own this which according to my theoretical calculations is 15X slower than Gigabyte 980ti extreme gaming.Where'd you see him owning, or mentioning, that 980 :hmm:
You are in a good spot.
If AMD is honest in saying, minimum VR specs at lower entry point cost, then you can get 2 Polaris 10 models for less that 1080 FE, especially as VR seems a great match for dual GPUs.
Basically a slightly slower Radeon Pro Duo for much less.
You are in a good spot.
If AMD is honest in saying, minimum VR specs at lower entry point cost, then you can get 2 Polaris 10 models for less that 1080 FE, especially as VR seems a great match for dual GPUs.
Basically a slightly slower Radeon Pro Duo for much less.
You are in a good spot.
If AMD is honest in saying, minimum VR specs at lower entry point cost, then you can get 2 Polaris 10 models for less that 1080 FE, especially as VR seems a great match for dual GPUs.
Basically a slightly slower Radeon Pro Duo for much less.
the problem with nvidia is that they self locked themselfs on a premium note...Nvidia did just like Roy Taylor said they would, come out with a chip that does not address TAM. The GP104 is 317 mm2 https://www.techpowerup.com/gpudb/2839/geforce-gtx-1080 and does not address a large market. Especially at a price point of $430 and $700 even though they claim msrp of $380 and $600 which all the Nvidia bought tech sites will quote. Good luck with an AIB selling at that msrp.
When Roy made his comments everyone started saying Polaris 10 won't compete, and maybe they won't compete with the 1080. I tend to look at his comments in reference to P10 and P11. AMD from a business perspective is looking to APPLE and back to school laptops, console upgrades and also mainstream desktop. Laptop market is larger than desktop. Consoles are also a large market and then add in Apple.
They showed what a P11 chip can do back in December so Laptop manufacturers could plan their new 2016 laptops, console manufactures can consider upgrades and Apple could also plan their next generation products. Manufactures need a long time to design a new product and want to show them off at Computex. The fact that GP106 appears late to the game is a significant business misstep.
Desktop mainstream is also available to AMD with the release of GP104 geared to higher end. How P10 and 11 perform with VR will be an important factor. If you can provide 90 fps for the current VR headsets with a smaller gpu at less watts and lower price then why buy a more expensive high end gpu which provides the same results. For the enthusiasts (small TAM) such as many on this site they will need to decide if spending $700 or more on a mid range card for maybe 30% performance improvement over a 980TI or Fury X at 1080P and probably less at 4K is a good idea especially when 2017 comes with even better gpu's. The 1070 is still a secret with performance and Nvidia probably has a clue that P10 is competitive with it and is waiting for AMD to tip their hand. I trhink that if AMD captures 35% of these sales it will be a great success along with capturing OEM sales for entry level desktops.
Computex and design wins should be an indication of how successful AMD will be with Polaris and Bristol Ridge offerings. Nvidia to this point have not indicated that they are ready with GP106 but they may have a few surprises in store. Time will tell but it appears from what we know at this time that AMD is looking at a larger TAM than Nvidia.
Escalation of commitment refers to a pattern of behavior in which an individual or group will continue to rationalize their decisions, actions, and investments when faced with increasingly negative outcomes rather than alter their course. The related term sunk cost fallacy has been used by economists and behavioral scientists to describe the phenomenon where people justify increased investment of money, time, lives, etc. in a decision, based on the cumulative prior investment ("sunk costs"), despite new evidence suggesting that the cost, beginning immediately, of continuing the decision outweighs the expected benefit. These decisions are irrational in their current context but in alignment with decisions and actions previously made, and they can be influenced by a variety of determinants and contexts. The dilemmas leading up to such decisions usually involve the prior choices no longer working or causing personal or group losses. While there are options to either cease current actions or continue on with them, neither has clear outcomes or is the obvious choice. Escalation of commitment then occurs when persistence is chosen over withdrawal.
On the other hand, VR being so new, we can expect dual GPU to be included from early as the old AFR problems simply do not exist.Dual GPU is a great match for VR at some point in the future. Right now there are 0 VR games that use 2 GPUs as far as I know.
If that ever becomes common, we will probably be on volta and navi.
Are there not going to have the potential for many Polaris 10 models? Why are so many limiting the performance to narrow slots. As in I expect Polaris 10 to be FuryX, or 390X, etc.
With good clock gating, small die, 14nm Finfet properties, architectural improvements, surely we might get a huge range of performance.
The top uncut model should battle GTX1070, others ranging down to highly efficient notebook editions.