I thought it would be interesting to take what someone from a related industry, in this case the CPU market, and see what he thought what a market should be doing. Though his lecture is specific in some ways, taken as a whole it can be easily applied to the console market. The lecturer, Bob Colwell, was the chief architect of the Pentium Pro up through the Pentium 4.  

The lecture can be found at http://stanford-online.stanford.edu/courses/ee380/040218-ee380-100.asx

He hits on a few key areas 

exponential trends are not sustainable
-POWER
-BUGS
-MONEY

Need to change the game so that when a wall is hit, you don't sink. (Nintendo)
It is hard to convince a successful company that they need to change for their own good
-failure on the other hand, is a wonderful motivator for change.

The Public is weird
-you are not the public


Pick your targets judiciously
-don't go too broad

Be careful who you get involved with
-make sure your market is sustainableWhat platform has survived in a small, lucrative, and vertical (not expanding) market?

Don't plan to sell your product first - take it to the anvil and hammer on it until you get
something you want.

If you don't you have committed to features that you may regret in the future, and will have
to support in the future.

Benchmarks are easy to cheat on - be careful when looking at them.

Evaluate features fairly
-Don't spend a large portion of your budget on one feature with little benefit

Figure out how long you expect for your project to succeed and become successful - be explicit
-Don't bury your head in the sand.
-Ability to anticipate the future is limited - design accordingly

Standing behind your product when a huge screwup is shown can be a net-win.

Don't take business risks for other companies with no benefit to yourself

Most companies have a planning horizon of 3-6 months

Continuing to apply exponential ammounts of money to problems that people do not see exponential
gains to is not good business - its just a bad busines habbit.

technology is really the second priority to marketabilityBlue crystals - companies want things that either cost the same or less to market as "extra
value"

Marketing - why would you pay more for a lexus when its still a toyota. Marketing ploy
-Always want three products in your product line.
-Wendy's sold singles, doubles, triples - eliminated tripple because it
didn't sell. So eliminated triples, but lost doubles sales. Why? People
think that there is too little value for the single, think the double is ok,
and the triple is the pig out item. So they take the double.

-Important to give enough options to give appeal, but not too many to cause
confusion.  

  

Standing behind your product when a huge screwup is shown can be a net-win. Sony is doing precisely that behind the PS3. Whether it will be a net-win remains to be seen, but they're forging ahead on their strategy. 

Lingyis said: Standing behind your product when a huge screwup is shown can be a net-win. Sony is doing precisely that behind the PS3. Whether it will be a net-win remains to be seen, but they're forging ahead on their strategy.

I think Sony is standing behind their PAL customers more than the PS3. :P (sorry, couldn't avoid it)

 

In most engineering fields there is the concept of gold plated requirements and it is always seen as a bad thing ... The general principle is that you can continue to add or enhance features of a system in an attempt to build the perfect system, but you soon get to the point where the added cost is far smaller than the added value to the consumer.

Both Microsoft and Sony are guilty of gold plating the requirements on their home consoles. By now it should be clear that Sony and Microsoft could have released their console with a dual core PowerPC processor (probably a modified PowerPC 970MP), a modest gpu (Geforce 6 or Radeon x800 range), DVD drive, small ammount of built in flash (2GB) and optional hard drive and sell it for $300 and the average consumer would be happier for it.

Actually, the PS3 has less features than Sony wanted it to have, initially. I remember the time when the PS3 was also going to be your home router:

http://www.afterdawn.com/news/archive/6626.cfm 

 

Intel doesn't sell very much directly to the consumer. Their business model isn't really comparable to the console market. The pricing structure is completely different.

Fishyjoe: Listen to the lecture if you want a full explanation, but I'm assuming you won't so here is a quick run down of it.

 The people who buy intel products are the people who walk into circuit city, fry's, best buy, dell, etc. Intel has to sell a product that they feel the need to buy, or they lose. Think about it, why would intel show adds with blue man group with absolutely zero technical information to sell to OEMs? 

 Intel most certainly does sell their products to the public, and they are a business that makes their money by selling mass quantity, not by selling a million units a year total.

 As for the pricing structure comment? What do you mean? Are you trying to imply that just because one company has a more divercified product line that the lessons from their knowledge are somehow not adaptable to other markets? That simply just doesn't make sense. If you can find pause to disagree with any of the things I point out for the console market, thats one thing, but a dismissive blanket statement with no actual input is pointless.

Is this the same Intel that doesn't have a processor in any of the current major consoles?

You mean XBOX? Ya that Intel.

Yup, the Xbox had the Intel, until they were dropped.