Merrill Lynch's Next-Gen console prediction

[g35er]

In the end MS supporters will always agree with such articles regardless of the validity of them.
The article is not shameless or biased because it happens to support MS and you happen to agree with them and it makes you feel better to think that Sony iz teh doomed. The article is shamed and biased because it fails to address the advantages Sony has, which you have completely ignored (unless you think that technology is the only advantage Sony has, which in itself is ridiculous).

The article does seem biased. As mmmkay said, ML stated themselves that they could be biased.

Originally Posted by Merrill Lynch
Merrill Lynch does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report.

But for all the reasons people have stated in this thread, they can't just make up the numbers. They have to have justification for the cost numbers. You have to seperate data (the cost estimates) from opinion (interpretation of the numbers). Most of the discussion here is on the data, not on their opinion. Now, you can say they used "special math" to bias the numbers in favor of one comapny, but without knowing their sources, it's just speculation. We only know that they have public liabilities to get their data right or at least close.

 

[expletive]

They are still in the end predictions based on the information at hand, there's no ground for defamation. This report seems to be a private analysis which was probably not meant for us or even the media. If you haven't noticed yet, this report was leaked onto a forum and has only spread onto blogs and indie news sites which deal with user submitted info. Fairly effective marketing if you ask me.



Estimates and predictions are not facts. I said that it is quite likely their source information was based in fact but their subjectivity and spin (in my opinion) damages their credibility.

I was replying to someone who put this report in the same league one which gives recommendations to their investors. This report clearly does not and offers disclaimers to that degree. Would those disclaimers be on a report which does give recommendations?

What's the adage? If it's too good to be true, it probably is.

I didnt see the part where you said these are likely based on facts so apologies for that. However, when a company says theyve seen the BOMs and are using them to form opinions, I have to believe theyve seen both BOMs in a state where they can provide analysis. If they were to turn around and say 'we never really saw any BOMs' then i'll tkae my medicine and be the most surprised guy in the room.

 

[expletive]

I did. I was going to get PS3, but now thanks to ML I'm not going to save up for a machine that'll from a company that'll be bust in a couple of years. I'm waiting 'til Christmas next year when I'll get an XB360 for £150 including HDD, and not worry about exclusive games which I want to play on PS3 because this report didn't mention them so I guess they're not a factor anyone wants to consider.

Actually I'm lying. The report just said 'if this, and that, and we're right with these numbers we guessed at, then this might happen, which might be good for MS.' No different a 4 or 5 threads a month we get started here with people saying IF. Can't see the point of this report myself. Why did they bother to right it? Who's the audience and why'd they want to read it? I'm sure anyone can write speculative articles taking a number of IFs and presumptions to show how some company or other could out-do another. Guess they're just in a slow period and someone knocked it up in an afternoon when they had nothing better to do.

Cause thats what these firms do. They either get paid for an analysis or they just do them in the hopes that they are right enough so that they become the wall street authority on this stuff. I worked for one of these firms during the internet boom and we had an analyst on staff who was known as the master of technology IPOs, what they would be worth, who should buy, etc. She had a nickname and was renknown throughout. Thats the kind of thing these firms love, hacing the 'rainmakers' on their staff so they get notoriety and ultimately, rich people's money.

 

[Bobbler]

I didnt see the part where you said these are likely based on facts so apologies for that. However, when a company says theyve seen the BOMs and are using them to form opinions, I have to believe theyve seen both BOMs in a state where they can provide analysis. If they were to turn around and say 'we never really saw any BOMs' then i'll tkae my medicine and be the most surprised guy in the room.

BOMs might not always have costs associated -- they are mostly used for knowing what actually needs to go into the product. You often use a BOM to find out costs (because you know exactly what and how many), but a BOM itself doesn't necessarily have to have a cost of each part listed. They might have seen a BOM, but I'd bet my left nut that they didn't see a BOM sheet with costs on it (at least not for the big things, like Cell, RSX, Xenos, Xenon, BR, etc) or any other equivalent sheet like it.

 

[avaya]

A PS3 priced at total average cost is unrealistic IMO. It’s assuming they abandon the price equivalency to marginal cost less a subsidy, which has been the normal business model for all major players in the console market for a very long time.

This set of researchers at Merrill (Osha, Rangan, Ho & Post), not Merrill itself, are assuming that Sony will adopt a different business model based on comments made by Stringer:

CEO Howard Stringer has made some public commitments regarding the target level of profitability for the March 2007 fiscal year

OK fair enough, that’s for the whole Sony group I would imagine….

More than 75% of Sony’s revenues come from the CE division. The marginal effect of what SCEI do in relation to the rest of the Sony group is not going to be overly significant in that light. Proposing a complete change of a successful business model for SCEI based on those comments from Stringer is stretching it – indeed a worst case scenario.

 

[Mmmkay]

However, when a company says theyve seen the BOMs and are using them to form opinions, I have to believe theyve seen both BOMs in a state where they can provide analysis.

Err, they estimated all of the BoM calculations themselves.

Just for perspective, the source who leaked the Merrill Lynch article also put up one by citigroup. This is their cost analysis:
http://webpages.charter.net/spartan85/Next gen consoles_CITI.pdf


[edit]Ignore that last sentence. A brainfart just occured.

 

[Shifty Geezer]

Holy Cow! They've got Cells $188 a piece beginning of next year when these are going into the first PS3s. Someone ought to ring KK up and tell him he can get them cheaper if they bulk buy OEM, instead of buying over-the-counter chips one at a time in pretty packaging.

 

[mckmas8808]

More than 75% of Sony’s revenues come from the CE division. The marginal effect of what SCEI do in relation to the rest of the Sony group is not going to be overly significant in that light. Proposing a complete change of a successful business model for SCEI based on those comments from Stringer is stretching it – indeed a worst case scenario.

Great point and is something that I was trying to figure out why Merrill didn't take that into affect. I swear if some of these guys here were to talk through their NDAs were could get a close estimate too. With all the information we get here it shouldn't be that hard. As a matter a fact the new thread that one posted with the charts is also very help full in estimating a cost.

 

[mckmas8808]

Holy Cow! They've got Cells $188 a piece beginning of next year when these are going into the first PS3s. Someone ought to ring KK up and tell him he can get them cheaper if they bulk buy OEM, instead of buying over-the-counter chips one at a time in pretty packaging.

That's a crying shame. They have the PS3's total cost after everything at $658.

 

[expletive]

Holy Cow! They've got Cells $188 a piece beginning of next year when these are going into the first PS3s. Someone ought to ring KK up and tell him he can get them cheaper if they bulk buy OEM, instead of buying over-the-counter chips one at a time in pretty packaging.

Maybe that price includes a retail heat-sink and fan.

 

[Carl B]

Err, they estimated all of the BoM calculations themselves.

Just for perspective, the source who leaked the Merrill Lynch article also put up one by citigroup. This is their cost analysis:
http://webpages.charter.net/spartan85/Next gen consoles_CITI.pdf

Crack addicts, all of them. I don't even know what I'm looking at now. 'R&D and other overhead,' isn't that something that should be accounted for in the division financials, rather than a console BOM? And Cell is *still* appearing as more expensive than RSX there, throughout the life of the console. It defies all reason. For god's sake RSX will be the larger of the two!

At least it helps put down the notion of a solid BOM in the Merrill case, since they're both so wildly divergent.

 

[Dave Baumann]

At the moment the two parts are fabbed at different companies with different expertise. When it becomes integrated who'd do the fabrication? Isn't it TMSC and NEC? Are TMSC okay with eDRAM? If they are, why are NEC given the job of the eDRAM part? Or are there moves to advance the TMSC lines to support eDRAM?

Well, the other question is - why didn't NEC get the logic portion?

The current solution is probably just representative of where they felt th fabs were with each of the solutions given the timescales they were shooting for. If I were to take a guess then I'd say if/when it does happen TSMC will get the final design; if you take a look around the TSMC site you'll see that they have solutions for eDRAM on each of their major nodes, with 65nm planned as well.

 

[Carl B]

Well, the other question is - why didn't NEC get the logic portion?

Well I think that's an easy one to answer though; NEC fabbing = $$$ (comparatively)

They got the portion of the business that MS had to send there.

 

[Phil]

In the big picture, Sony has advantages and MS has advantages. This ML report just focus on the financial aspect, which they are well-qualified to do, and MS just happens to have the financial advantage.

I thought we've already established that they stuffed up their focus - the financial aspect - as seen by them using ludicrous numbers to form their "financial analysis".

How much is such a financial analysis worth, if they fail to look at the bigger picture (technology, the fabbing process, CELL compromising of redundant structures etc) that would change the result of the financial analysis greatly?

In other words, based on the fact that CELL, although larger in die-size, has the advantage of it being used even with defects to a certain degree and the point that defective CELL chips can be used for other products, I for one would be carefull to weigh CELL to be more expensive than Xenon. Considering that, why even take claims of $$$ for chips that aren't taking these factors into account even serious? Certainly doesn't sound "well qualified" if the average B3d member can come up with better estimates based on these few (well known) established facts...

 

[Deepak]

This story is on Yahoo now.....

http://news.yahoo.com/s/mc/20051103/tc_mc/merrilllynchuserslikelytopreferxbox360overps3

 

[Powderkeg]

How much is such a financial analysis worth, if they fail to look at the bigger picture (technology, the fabbing process, CELL compromising of redundant structures etc) that would change the result of the financial analysis greatly?

What makes you so certain that they didn't?

 

[Carl B]

This story is on Yahoo now.....

http://news.yahoo.com/s/mc/20051103/tc_mc/merrilllynchuserslikelytopreferxbox360overps3

Well of course, it was only a matter of time anyway. Though didn't think it'd make it to the 'Yahoo' level so fast.

 

[Carl B]

What makes you so certain that they didn't?

Well it would cost less than RSX if they had, that's for sure.

 

[avaya]

'R&D and other overhead,' isn't that something that should be accounted for in the division financials, rather than a console BOM? And Cell is *still* appearing as more expensive than RSX there, throughout the life of the console. It defies all reason. For god's sake RSX will be the larger of the two!

At least it helps put down the notion of a solid BOM in the Merrill case, since they're both so wildly divergent.

They think it will be smaller than Cell...

As with the Cell, we understand that the RSX will be initially manufactured using a 90nm design rule, with subsequent migration to 65nm. We estimate chip size at around 210mm2–240mm^2.

Cell....

First, the PS3’s Cell CPU is made from 12-inch wafers and uses 90nm processes, and we think initial production phases will entail considerable volumes of dummy wafers. Next, we assume the effective usable area per wafer at 57,000mm2 having taken edge exclusion (the narrow band on the outside of a wafer cannot be etched with a circuit pattern) into consideration.

As we assume the Cell chip size at 235mm2, we calculate that around 243 chips can be derived from each wafer. While yields need to be additionally considered here, we forecast initial phase yields at around 30% as logic yields are typically low when lines are newly started up. We thus take 30% of an assumed 243 chips on a front-end process basis in arriving at an estimate of around 73 good chips.

The back-end process comes next, and we forecast yields of around 75% during packaging and testing processes. We assume the number of chips derived will thus further decline from the 73 good chips yielded from front-end processes, ultimately amounting to 55 good chips.

We calculate the final cost per chip to be $329 after aggregating processing costs, packaging costs, and testing costs, and dividing this total by 55, the number of good chips.

EDIT:

NikkoCitigroup are saying XeCPU won't be using SOI. This is fatcually wrong, they say the use of SOI negatively contributes to Cell's defect rate.

 

[Carl B]

They think it will be smaller than Cell...

Right... but why would RSX on the same process with roughly ~70 million more transistors than Cell actually be smaller? And going further, with their yield estimates they're completely ignoring the fact that die defects to a certain extent can be tolerated on Cell, whereas their yield estimate numbers seem to reflect the percentage of perfect dies, as would traditionally be the case.