Rumor: Price Cuts on GTX660Ti series coming next week

[blastingcap]

Granted, I don't live in Hawaii but going by that chart above (it doesn't state specific values above the >15 cent point) I can only say I'm as screwed as they are without the beautiful weather and gorgeous scenery. Haha.

Hawaii = something like 32-44 cents per kWh for residential rates, depending on which island you're on. But yes they get the weather and scenery. And better weather also means less need for a/c and heating, which helps offset the high cost of electricity.

http://www.heco.com/portal/site/hec...mt=default&vgnextrefresh=1&level=0&ct=article

 

[RussianSensation]

GTX660 pre-order pricing:

PCSuperStore.com – $289
CostCentral.com – $289

Is this card going to be $199 or $229?

 

[RussianSensation]

Yes, it makes sense. AMD's revenue was down from Q1 -> Q2. nVidia's geforce revenue jumped 15% from Q1 -> Q2. nVidia said that they are even supply contraint in the high end because of the demand. There is a reason why AMD cards lost nearly 30% of value in the last 4 months.

:hmm:

Using AMD's total revenue vs. NV's total revenue to extrapolate sales of 1 specific product segment, i.e., desktop discrete GPUs is finance fail 101.

- AMD sells CPUs and GPUs. AMD's product segment cash flows currently break down roughly as follows:

Notebook CPUs: 26%
Desktop CPUs: 20.6%
Server CPUs: 13.5%
Graphics GPUs: 23.5% (11.2% Discrete Notebook GPUs, 8% desktop discrete GPUs, 4.3% Professional gaming & consoles)
Chipsets: 12.1%
Embedded CPUs: 4.1%

Desktop discrete GPU segment for AMD only represents approximately 8% of the company's cash flow. (i.e., represent less than 1/12th of the company's value).

Now Nvidia:

NV has:

- Professional GPUs
- Discrete desktop and notebook GPUs
- Mobile and game console chips
- PC microprocessors

Desktop discrete GPUs for NV are projected around 17-18% of the company's cash flow. (i.e., represent less than 1/5th of the company's value).

You cannot compare total revenue of both companies to how the companies are performing specifically in the desktop discrete GPU segment. That's not even remotely logical. :sneaky:

Funny how you actually believe that GTX680 sold more units worldwide than HD7750 or 7770? good one!! (Only around 3-5% of consumers buy GPUs > $400).

The reason AMD dropped prices is because they appear to have used the First Mover Advantage strategy in business that allows a company to dictate prices until the competitor shows up. The new price levels also are what ATI used historically. This is just going back to the norm instead of selling GPUs at low margins. By having a technological leadership in the marketplace temporarily, you can dictate higher prices and make pricing adjustments only when necessary (i.e, when the competitor launches). This is no coincidence.

If you look at the prices AMD has settled at now $299-319 for 7950 and $419 and $449 for 7970/7970GE, those are still higher than $269/369 (5850/5870) or $299/$369 (6950/6970). The risk is that NV can beat AMD to market or release much faster product which would force AMD to revert back to price/performance.

What is amazing is that AMD still sells a faster GPU for less $ (7970GE vs. 680). So if anything it is NV that dropped the ball since they have allowed AMD to raise prices by providing less competition than during HD2900-6900 generations. In each of those, the flagship NV card had a 15-20% lead. This generation, 680 is not the fastest GPU.

 

[l2ez4m]

Found one Asus 660 in EU, which does come at exactly 230$ after usual in EU 20% VAT deduction.

 

[RussianSensation]

Statistics is fun! You can phrase numbers to represent whatever you want!A random web survey shows Company A selling 9 times as many units as Company B. OMG web survey must be useless/biased!

Except we know exact AMD and NV discrete GPU market share. So your example doesn't work based on real world market share data. AMD gained discrete GPU market share last quarter, Steam survey says otherwise. Therefore, Steam Survey is worthless as has been shown for many years and people still keep linking to it. Steam Survey has already failed miserably before when people claimed Fermi cards weren't selling well because Steam Hardware Survey was completely dominated by HD5700/5800 series. If companies choose to use Steam for marketing, and people buy the data, that's their own call. Real world data has proven that Steam Survey is just a guestimate and more often than not is way off. There are plenty of gamers in large markets (BRIC countries - Brazil, Russia, India and China) and many of those gamers do not use Steam at all or don't even have Internet. Since BRIC countries by far represent the largest growth of PC gamers worldwide, that makes Steam Hardware Survey even more worthless for real world market share / sales data.

 

[acx]

Except we know exact AMD and NV discrete GPU market share. So your example doesn't work based on real world market share data. AMD gained discrete GPU market share last quarter, Steam survey says otherwise. Therefore, Steam Survey is worthless as has been shown for many years and people still keep linking to it. Steam Survey has already failed miserably before when people claimed Fermi cards weren't selling well because Steam Hardware Survey was completely dominated by HD5700/5800 series. If companies choose to use Steam for marketing, and people buy the data, that's their own call. Real world data has proven that Steam Survey is just a guestimate and more often than not is way off. There are plenty of gamers in large markets (BRIC countries - Brazil, Russia, India and China) and many of those gamers do not use Steam at all or don't even have Internet. Since BRIC countries by far represent the largest growth of PC gamers worldwide, that makes Steam Hardware Survey even more worthless for real world market share / sales data.

People believe what they want to believe. My example was fake example. It has no relation to any real world data. It only points out that true statements about statistical data can be misleading when taken out of context of the entire data set.

Steam survey isn't worthless. It just doesn't measure what some people keep using it for. It is a sampling of the hardware that Steam users are running. Steam users are a biased sampling of the world population. Maybe that's not what you want but that doesn't mean the survey has no value at all.

 

[RussianSensation]

Steam survey isn't worthless. It just doesn't measure what some people keep using it for. It is a sampling of the hardware that Steam users are running. Steam users are a biased sampling of the world population. Maybe that's not what you want but that doesn't mean the survey has no value at all.

Ya, agreed. My point was that members here use Steam Hardware Survey as indicative of worldwide desktop discrete GPU sales. For that particular metric, using Steam Hardware Survey as an accurate estimate of global sales is a pretty bold assumption.

 

[SirPauly]

It offers some insight what gamers may be using considering the large data base that Steam is. It's not official information and certainly not the end-all-be-all metric and should be some caution when looking at the data. Official information utilizing Jon Peddie Research or Mercury Research are more ideal data resources when it comes to shipped sku's.

The key though Steam adds fun discussion points and adds data points from a large gamer resource monthly and breaks down the sku's.

 

[3DVagabond]

No it's not! It offers some insight what gamers may be using considering the large data base that Steam is. It's not official information and certainly not the end-all-be-all metric and should be some caution when looking at the data. Official information utilizing Jon Peddie Research or Mercury Research are more ideal data resources when it comes to shipped sku's.

What's the big deal? These are only video cards -- no need to be emotional.

Steam survey could be as dodgy as a Russian election, for all we know.

 

[JTsyo]

So when can we expect the news if in fact it's happening this week? Tuesday?