Ed Finnell writes: >The beauty of the Z6/Z10 was that by using the G6 >Power fabrication foundry IBM should be able to use >a single supply chain for materials and packaging.
I would beg people to just go to Google, find the "30,000 foot" digital images of the z10 and POWER6 microprocessors, and compare them side-by-side. They look completely different because they are extremely different. Now, among the similarities: 1. They both have binary architectures, operating on zeroes and ones. 2. They both have 64-bit addressing. 3. They both execute instructions (adds, branches, etc.) 4. They both run on electricity. 5. They both contain silicon and certain metals. 6. They both are installed inside servers. 7. They both can access memory. 8. They both have processor cache memory.. 9. IBM manufactures both, and sometimes even in the same building. 10. IBM ships both of them from their birthplace(s) to other buildings, using moving vehicles. 11. Engineers who design them often talk with each other, even cordially, occasionally (but not exclusively) over a beer. Maybe two. And there may be other similarities, too. For example, they both have IEEE754r hardware decimal floating point units. They are both quad-core. They both currently run at clock speeds well in excess of 4 GHz. I really, really don't know where this "they're the same" mythology got started. Take a look at Charles Webb's presentation if you want a partial list of the many differences. Or just look at the photographs! But who the heck cares? What I care about is that a processor delivers and continues to deliver the most highly favorable execution qualities for my business applications (and the operating systems and middleware that support those applications). Hardware is a doorstop without good software,(*) and we should evaluate hardware based on what business benefits (via software) it delivers. If there's something nifty in a POWER processor or a YUNKLE processor or a TurboPLONK processor that could be helpful if incorporated into a System z processor, great, I'll happily take it. I don't particularly care where it comes from. [Note: YUNKLE and TurboPLONK are not real processors as far as I know, although I heard IBM is working on SuperTurboPLINK. :-)] And, to be clear, "most highly favorable execution qualities" encompasses reliability, availability, serviceability (RAS), security, execution and data integrity, etc. All the "-itys." (*) Although if it's System z, it's a very beautiful *and* effective doorstop. :-) Wheeler writes: >After worldcom/enron ... GAO started a study of such >reporting and looking at statistics of "refillings" >... i.e. were subsequently reporting is refiled because >of (at least) audit and accounting errors. This would be the "I don't believe the number is accurate" argument I guess. Well, if you think there's an error in IBM's earnings report, you can contact the SEC by visiting http://www.sec.gov. Ted MacNeil writes: >Just because the CFO said it [System z gained marketshare],, >it doesn't make it so. To prove an increase in share you need >at least a market size to compare it to. This would be the "I don't believe +32% is big enough" argument I guess. I don't believe +32% is big enough either, but I'm never satisfied. :-). Yes, you do have to look at the total market size. Presumably IBM's CFO has a good idea of that total market size, and I can understand why. All you have to do is find some respected and competent server market analysts who determine that the entire ex-System z server market grew at least 32% last quarter. I think you'll find that's a very small group of analysts. :-) The reality is that "gained marketshare" is about the most *unremarkable* conclusion that could be drawn from such a stellar System z performance. Anyway, could we all just applaud and then talk about something else, like what USS stands for? :-) I mean, it's +32%, and thank you! And a very special thank you to our new System z customers, including the first customer in Vietnam. - - - - - Timothy Sipples IBM Consulting Enterprise Software Architect Specializing in Software Architectures Related to System z Based in Tokyo, Serving IBM Japan and IBM Asia-Pacific E-Mail: [EMAIL PROTECTED] ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
