Merrill Lynch's Next-Gen console prediction

xbdestroya said:
I want to reiterate my own belief that indeed the 360's component costs will come in below PS3's, and remain there for a couple of years, if not throughout.

And for the record to everybody reading (this means you jvd) I will second xbd. I'm not sure if it will be throughout but I most certainly think the 360 will cost less for a few of years at least.
 
scooby_dooby said:
I think many pro-sony guys are using the fact that Sony produces it's own chips as some sort of universal excuse that they can include any technology, no matter how expensive and somehow just because they have their own fabs that it's suddenly drastically cheaper to produce.

BR drives STILL have not even launched in North America, it's extremely expensive technology to manufacture right now. In addition, if BR doesn't win the format war Sony is stuck with a technology they cannot cost-reduce over time in a meaningful way, regardless of whether they have their own fabs or not.

The also have CELL that may or may not make it's way into consumer electronics(The EE sure didn't). If not, many of the arguments for Sony recouping losses fall to the wayside, and they're stuck with an expensive CPU.

ML pegged BR at $101 or so to produce. There's companies that specialize in nothing but estimating hardware costs, and I'm sure ML went to one of these companies for an analysis, and I'm sure that estimate is pretty close to the truth, not complete rubbish like many are claiming.

im surprised u realised just now how pro sony alot of b3d console-talk participants are
 
xbdestroya said:
The arguments I've made stand; my biggest problem is they claim Cell will cost more than RSX. It simply doesn't make sense.
OK, so I saw you link to this:
Will the RSX, if really a modified G70, be so much smaller than 221mm^2 on 90nm going from 334mm^2 at 110nm? No.
(110/90)^2 = 1.5. ==> 334/1.5 = 223.

Similar size. RSX will have high yields because it doesn't need to be clocked this high (bandwidth limitation), and G70 is already near this clock on 110 nm. CPU's have much lower yields. The more I think about it, the more I realize that the defining reason is reliability. RSX can have every billionth pixel be bad. Cell, or any other CPU for that matter, doesn't have this luxury. A random flaw can crash the system, and it runs at 3.2 billion cycles per second.

Anyway, if you're going to continue ranting about this, address the last paragraph of my previous post. The point is that the ratio of Cell cost to Xenon cost is what you should be focussing your argument on. And you really have nothing to complain about there.
 
i still believe that if this thread cast the Xbox 360 in a negative light, there would be much more agreement with ML...even by hardcore Xbox fans because even they are timid to defend the Xbox in a non pro-Xbox environment. I think because Sony had two relatively uncontested wins, they get a pass more often than not.

I seriously can't believe the number of "Well anyone who thinks Sony won't dominate next gen is stupid!" posts across the net considering the history of the industry and the fickleness of game consumers in the market. Previous installed base, big-named franchises, billions in the bank and brand recognition couldn't keep Atari, Nintendo and Sega at the top. In the end it was either one stupid choice or a series of stupid choices that allowed a strong competitor to overtake these previous market leaders.

I mean, who could have predicted that a seemingly insignificant move such as Nintendo breaking off a deal for a Sega CD rip-off from Sony would have resulted in Nintendo being relegated to a niche market, and the maker of "Mickey Mania" for the Genesis (Sony), would sell over 200 million consoles?
 
scooby_dooby said:
The also have CELL that may or may not make it's way into consumer electronics(The EE sure didn't). If not, many of the arguments for Sony recouping losses fall to the wayside, and they're stuck with an expensive CPU.

Do some more research. Matter of fact don't I will do it for you.


Mercury Computer Systems has created the Dual Cell-based Blade, a computer server capable of processing at 400 gigaflops thanks to two IBM Cell BE (Broadband Engine) processors. The Cell is a single-chip multiprocessor with nine processors operating on shared memory.
The Dual Cell-based Blade is expected to greatly boost the performance in products that rely on radar, sonar, MRI and digital X-rays for customers in aerospace, defense, seismic and medical fields. For example, military reconnaissance and surveillance groups might use the new blade to process radar, sonar, and signals. Customers might also use the blade to run medical diagnostic imaging devices including MRI, PET, and digital X-rays.

Chelmsford, Mass.'s Mercury said orders for the hardware, the first Cell-based product to market outside the gaming space, are expected in the first half of 2006. The blade system is the fruit of a June 2005 agreement, in which Mercury partnered with IBM Engineering & Technology Services to integrate Cell technology into products designed to address applications that require a lot of number crunching to run to their potential.
Link http://www.internetnews.com/ent-news/article.php/3554141

Toshiba has announced that the mysterious CELL processor, set to be used in the Sony Playstation 3 (PS3) console, will be in all their TVs in 2006 as well.

http://www.engadget.com/entry/1234000610026194/

And don't forget that Sony will be using them also. Well looks like the CELL is already pasted the EE as far as multiple usability is concerned. Bring on your next point.

CELL -> EE lesson learned.
 
Last edited by a moderator:
Mintmaster said:
OK, so I saw you link to this:

(110/90)^2 = 1.5. ==> 334/1.5 = 223.

Similar size. RSX will have high yields because it doesn't need to be clocked this high (bandwidth limitation), and G70 is already near this clock on 110 nm. CPU's have much lower yields. The more I think about it, the more I realize that the defining reason is reliability. RSX can have every billionth pixel be bad. Cell, or any other CPU for that matter, doesn't have this luxury. A random flaw can crash the system, and it runs at 3.2 billion cycles per second.

Anyway, if you're going to continue ranting about this, address the last paragraph of my previous post. The point is that the ratio of Cell cost to Xenon cost is what you should be focussing your argument on. And you really have nothing to complain about there.

Mintmaster have you read the thread? You seem not to know what I'm ranting about.

Here go see this: Link

We are using the G70's die size as a baseline for the RSX simply due to the fact that we expect them to be similar, but that said I don't expect them to be the *same*. Certainly there must be some accounting for a modified architecture vs it's predecessor fab-wise when the derivative will be on low-k and SOI when the original is not. Not to mention, a different fab once again.

Also I just have to repeat that what makes dies expensive other than die area, yield, and defect-rate? With the Cell able to withstand a defect now and then and still make it into PS3, I don't see why you're putting up such a wall to the notion that Cell could cost less than RSX.

Finally Cell will have been in production (though in limited quatities) for over a year by the time PS3 launches, and likely ~9 months before it goes into volume production. This is a chip that should already have the process reaching some state of acceptable yield by that point, rather than the 30% yield posited.

We can debate this all day long - I'm full well willing. My cost comparisons have nothing to do with Cell vs XeCPU. I've never said Cell would be cheaper than the 360's chip. What I'm arguing is the bogus BOM numbers here, and how a firm that can put RSX at $75 would put Cell at $248. Do you see my point? My case? It has nothing to do with 360 vs PS3; it's all about PS3 vs PS3 - costwise. These numbers are ridiculous.

I went off on the 360's $25 DVD drive as well earlier in the thread (the costs of that drive should be much lower), so I'm hardly limiting it to the silicon inconsistencies; they're simply the more egregious.
 
Last edited by a moderator:
Mckmass - Workstation are not consumer electronics.

As for Toshiba using CELL in every single TV? I'll believe that when I see it, I have no idea why they would bother including CELL in the vast majority of their TV's, what does a 32" SDTV require CELL for? It sounds extremely far-fetched.

And like I said, it MAY or MAY NOT happen(you seem to have missed the MAY and went directly to the MAY NOT), you can tell me to do research all you want, but the fact is there's nothing to prove your point since it hasn't yet made it's way into consumer electronics, so it remains to be seen what will happen.
 
scooby_dooby said:
Mckmass - Workstation are not consumer electronics.

As for Toshiba using CELL in every single TV? I'll believe that when I see it, I have no idea why they would bother including CELL in the vast majority of their TV's, what does a 32" SDTV require CELL for? It sounds extremely far-fetched.

And like I said, it MAY or MAY NOT happen(you seem to have missed the MAY and went directly to the MAY NOT), you can tell me to do research all you want, but the fact is there's nothing to prove your point since it hasn't yet made it's way into consumer electronics, so it remains to be seen what will happen.

What I'm trying to say is CELL will be used for more things then just the PS3. Hell the PS3 hasn't made its way the consumer electronic stores, yet you believe they will use it to power the PS3 though.:???: Point being Mercury Computer Systems has paid money to use the CELL for products that rely on radar, sonar, MRI and digital X-rays for customers in aerospace, defense, seismic and medical fields. Don't you think they have to work and test this stuff out in workstations before they release it to the market?

DeanoC and nAo right now are probably working on a workstation making HS, yet you knock Mercury Computer Systems and Toshiba? Toshiba is one-third of STI why wouldn't they use the CELL processor if they said they wanted to?
 
mckmas8808 said:
What I'm trying to say is CELL will be used for more things then just the PS3. Hell the PS3 hasn't made its way the consumer electronic stores, yet you believe they will use it to power the PS3 though.:???: Point being Mercury Computer Systems has paid money to use the CELL for products that rely on radar, sonar, MRI and digital X-rays for customers in aerospace, defense, seismic and medical fields. Don't you think they have to work and test this stuff out in workstations before they release it to the market?

DeanoC and nAo right now are probably working on a workstation making HS, yet you knock Mercury Computer Systems and Toshiba? Toshiba is one-third of STI why wouldn't they use the CELL processor if they said they wanted to?

But the crux of the entire "CELL will pay for itself" and "Sony doesn't need to have good yields cause they can re-use CELL" arguments rest in the major adoption, in a meaningful way of CELL in consumer electronics.

As for Toshiba, they will use whatever is most cost effective going into 2006. That link you posted, January 2005! It's now Nov, and I haven't heard anything about Toshiba's new line of CELL basd TV's. Aren't the 2006 models already rolling into store shelves?

Companies say alot of stuff, it doesn't always happen.

My point is many of these arguments are based on very favourable "what-ifs", CELL being adopted for CE's and mass produced is a huge IF. And Blu-Ray winning the HD format war is also a big IF, although it's looking more and more likely, the first players have not even reached our shores, so it's a far cry from a slam dunk at this point.
 
Alpha_Spartan said:
i still believe that if this thread cast the Xbox 360 in a negative light, there would be much more agreement with ML...even by hardcore Xbox fans because even they are timid to defend the Xbox in a non pro-Xbox environment. I think because Sony had two relatively uncontested wins, they get a pass more often than not.

What Ive learned here on the B3D Console forum is that generalizing opinion in such a dismissive and skewed manner (as you've done here) isnt usually the best way to promote conversation and civil debate. Disagreeing with a general sentiment without providing any nuance or addressing individual points by lazily phrasing the cadence of discussion as "pro Sony" or "pro MS" is simply false martyrdom, and very boring at that.
 
PS3vsXbox360.jpg


Wow now that's a big difference of cost.:oops: Ken needs to just toss the XeCPU into the PS3 now before Sony loses the next-gen war.:D
 
Last edited by a moderator:
I think he makes a very valid point though.

If it had been an anti-MS article it would not have been torn apart and attacked like this article was. Probably would've gone 2 pages and slipped off the front page...
 
mckmas8808 said:
Wow now that's a big difference of cost.:oops: Ken needs to just toss the XeCPU into the PS3 now before Sony loses the next-gen war.

Still waiting for you to show me why Toshiba is not releasing CELL based TV's since 2006 is less than 2 months away...

If you're going to make comments like "Go do some reasearch" which I find very insulting, at least have the cahones to genuinely debate the facts. Where are these CELL based TVs? Why haven't Toshiba announced anything for over 10 months?
 
Mckmass I don't get it - why do you post stuff like this, say something like 'Sony should go with XeCPU,' and then undoubtedly spend the next twenty posts arguing with people who might suggest that very thing? ;)

That chart's just the visual representation of those ridiculous other numbers though, so nothing new.
 
scooby_dooby said:
Still waiting for you to show me why Toshiba is not releasing CELL based TV's since 2006 is less than 2 months away...

If you're going to make comments like "Go do some reasearch" which I find very insulting, at least have the cahones to genuinely debate the facts. Where are these CELL based TVs? Why haven't Toshiba announced anything for over 10 months?

That's all I seen on the website. My point to you was that CELL will make it into more products than just PS3. I noticed that you said maybe or maybe not. I just didn't like the little EE remark. You implied there that CELLs outcome could be like the EE, which are in two completely different leagues. I didn't see the need into comparing the two. That would be like me comparing the Xbox to the Xbox 360. I would never say that just because the Xbox didn't turn a profit for MS that the X360 probably wouldn't either.

Two completly different beast you see?

xbdestroya said:
Mckmass I don't get it - why do you post stuff like this, say something like 'Sony should go with XeCPU,' and then undoubtedly spend the next twenty posts arguing with people who might suggest that very thing? ;)

That chart's just the visual representation of those ridiculous other numbers though, so nothing new.

Just having a little fun thats all.:D I should have placed a smiley at the end. Maybe I'll do that now...
 
mckmas8808 said:
PS3vsXbox360.jpg


Wow now that's a big difference of cost.:oops: Ken needs to just toss the XeCPU into the PS3 now before Sony loses the next-gen war.:D

Is there any point in putting the HDD on there when it won't be included with the PS3? It seems to me the best comparison would be to compare the PS3 costs with the costs for the core 360 console.
 
Branduil said:
Is there any point in putting the HDD on there when it won't be included with the PS3? It seems to me the best comparison would be to compare the PS3 costs with the costs for the core 360 console.

I believe they are referring to Blue Ray vs. DVD.
 
And people lets get back to the basics here. Do we here at B3D expect Sony to really launch for $499.99? Of course not. I think the PS3 will be $399.99 maybe even $349.99 with less price drops.

I look at it this way say the PS3 is $500 who many people will buy one? How many people will buy millions of games for Sony like they have in the past? With less consoles sold there will be less software sells. And if the PS3 will be twice the price of the X360 next holiday, then what would stop MS from dropping the price to $149 when Sony drops the price of the PS3 to $300? To me that could possibly mean that MS could have the X360 at $99 4 years from now while Sony is just be getting to the $199 mark (a price that MS would have hit 2 or 3 years before).

How many people will download content from Sony:Connect due to the fact that they won't be buying a PS3 due to the outragoues price difference between it and the 360? How many people are going to buy Blu-ray movies now that they PS3 is $500 compared to if the system started at a price point between $299.99 and $399.99?

If you look at it that way, then you would see that $500 would kill their only cash cow. Sony's other divisions could never bank off of the PS3 due to it's horrible price disadvantage. Blu-ray wouldn't catch on no where near as quick as the movie houses are hoping it will. Sony's movie division wouldn't as much as it could if the PS3's price wasn't comparable to the 360.

Sony's electronics division would be hurting due to the lack of economy of scale. The PS3's two biggest advantages (i.e. Blu-ray and CELL) could bust. If that happens with the lose of the console market advantage that they have had for the last decade wouldn't Sony be in a worst situation than they are in now?

Why would Howard Stringer praise over and over about putting the PS3 in the middle of everything Sony related if they can't even price the PS3 comparable to the other two next-gen consoles. Yes I didn't forget (maybe you guys did) that the Nintendo Revolution will cost even less than the Xbox 360. So while the Xbox 360 could be $250 next holiday season the Revolution could be $199.

So what do we have here? Holiday season 2006 we have

A.) PS3 for $500 (without a HDD)

B.) Xbox 360 for $250 (with a HDD)

C.) Revolution for $199 (with 512MB of Flash memory)


I'm sorry I just don't see that happening.:???:
 
Back
Top