I seriously hope that the R700/4870x2 has a better cooling solution than the 4870x1 has...these things sound like leaf blowers and the cards still
run hot, even with 7 case fans...then again, I hear this is somewhat due to immature catalyst fan drivers for the 4800 series.
This article is about ray tracing (i.e. rendering scenes as used in SFX and 3d animation applications), not gaming performance. So, your description of the quoted specs as "playable" is a bit inappropriate. Although the addition of raytracing compatibility (especially at the quoted performance levels) in ATI cards is a great addition, especially for those in the film industry, it doesn't mean much for gamers or general PC users at this point. Very few consumer 3d graphics applications (Maya, C4D, Blender, whatever) implement this type of GPU acceleration at this point anyway, so unless you're in the extreme-high-end 3d graphics business and are coding your own rendering software, this is pretty much irrelevant.
...although it is nice to know that ATIs market share will be bolstered by purchases from the extreme-performance segment of the SFX/film industry
|Originally Posted by a1161979
And prehaps you fail to realise that being an oligopoly AMD will not price and supply at the intersections of the Marginal Revenue and Marginal Cost curves and therefore your supply and demand idea suddenly seems a little silly Prehaps you should actually study economics before making such claims. Clearly considering the current market situation for both Ati and Nvidia it is not in eithers interest to interact with the market in a supply and demand sence
First of all, oligopolistic firms DO produce a quantity
at which their marginal revenue equals their marginal costs; any firm (monopoly, oligopoly, or perfect competitor) in a market economy will. Oligopolistic forms may charge a price higher than what their marginal costs require (due to a demand which exceeds their marginal revenue), but the quantity or "supply" will not be changed over any other type of industry.
Secondly, the effectiveness of price inflation due to the demand-exceeds-MR characteristic of an oligopolistic market requires all of the relevant firms in the industry to exhibit collusion (commonly called "price-setting") and therefore agree to price within a Nash Equilibrium model. Collusion is illegal
;-) and an industry in Nash Equilibrium tends to fail within a short time anyway due to the incentive for firms to cheat. The graphics card industry is most certainly not
in NE, and the two leading firms are non-collusive. ATI/AMD and nVidia operate as independent
(non-collusive) oligopolistic firms whose products are in valid price competition; ATI is perfectly able to produce at a level at which their price-to-performance ratio is lower than nVidia's, but customers will still be free to buy the substitute good (i.e. an nVidia card) if given financial incentive to do so. As a result, both companies are
affected by the market forces in the industry, just as if they were in monopolistic competition rather than oligopolistic competition, and are
subject to the pulls of market supply and demand.
The only difference between the current state (noncollusive oligopoly) and if the industry were in perfect competition is that both producers (ATI and nV) may charge at prices above their marginal costs, therefore earning a profit. The extent to which they can do this, however, is dependent on their demand curve, which in turn is dependent on the actions of the competing firm. If (as you suggest) ATI were to choose to sell their products at a higher price, they would suffer a loss in total revenue and profits due to the availability of a substitute (nVidia) which affects their own demand. Both firms are fully aware of this, and therefore DO price their products as competitively as possible; if the VGA industry were truly formed as you suggest it to be, the two companies would be able to charge exuberant amounts for graphics cards and not suffer due to perfectly inelastic per-firm demand; this is most certainly not
the case. While a two-firm industry (more accurately named a duopoly
) has a perfectly inelastic aggregate
demand, each firm has a limited demand curve (not perfectly inelastic, perhaps only slightly above unit elastic) which is dependent both on the other firm's competing products and the market health in general. There are even other firms (albeit smaller ones like Intel and Matrox) in the graphics card industry, which makes it more of a monopolistic competition than an oligopoly anyway; new companies may freely enter the market provided that they can produce products which rival the leaders' performance-per-dollar. Intel has already shown with their integrated graphics chipsets that this is possible, and are seeking to enter the mainstream and enthusiast graphics market soon (see here)
Furthermore, the 4870x2 (or R700) has its own market model which is not
identical to ATIs as a whole. So, not only is it in price-to-performance competition with nVidia's competing products, but also with ATI's other graphics cards.
If ATI chooses to charge $800 for the R700, consumers will gladly purchase 2 4870x1 cards instead.
In short, demand affects any
type of firm in a market economy; the fact that oligopolistic firms can earn a bit more profit due to having a larger share of the market is irrelevant; the market is still a market. If there was zero demand for graphics cards, both companies would go out of business just as quickly as if there were a million firms in the graphics card industry. If one firm decides to start charging a $500 premium on all of their cards, the consumers can easily choose to buy from the other firm instead. ATI and nVidia both maximize profits by producing and pricing at a point determined by their respective production costs (a firm's "supply curve") and their respective demand, which is derived from both the market's demand in general (i.e. how many people want/would buy a graphics card in the first place) and their competitor's substitute product. A particular product (say the R700) is also subject to substitution-effect demand elasticity from the same producer's other products (ATI's other graphics cards). So yes, ATI could
choose to charge an exuberant amount for the R700, but the would-be buyers will simply buy GTX280s or some other ATI card (4870s in crossfire?) and ATI would lose money on the R700. ATI isn't stupid and won't willingly choose to make less profit by ignoring market demand.
I seriously doubt your claim to being an economics student, as anyone who has been through even a high school microeconomics class would know this. Please refrain from attacking the casual remarks of others unless you want it for yourself, and don't start boasting about your would-be qualifications that are totally irrelevant to the topic. Go troll somewhere else, please.
|Well i thought i'd cut down the guy who thought he knew some economics
...irony? Edited by CiDaemon - 7/6/08 at 12:19am