111–120 of 159 posts
Server CPUs are way too cheap compared to the remaining parts of the server. Paying 10% more in exchange for software compatibility is a steal.
Earlier quoted context omitted.
http://www.anandtech.com/show/9959/samsung-announces-14lpp-m... https://semiaccurate.com/2016/09/01/intel-finally-narrows-14...
I think you might have forgotten to type a post to go with those links. Later iterations of Ryzen are undoubtedly going to fix some of the teething issues (segfaults, memory clocks, etc), and probably get the clocks up too. But they will also be going up against Coffee Lake and others - Intel can iterate on their products too. Consumer Ryzen is basically just Haswell with an AMD sticker on it, and it's just so transp…
Earlier quoted context omitted.
Available software constraints choices in many of those cases. Put two choices in front of an application team exec: rewrite and port an application from POWER/AIX, POWER/System i, POWER/System z for $X million, with all the attending risks that entails. Or spend same $X million on new IBM hardware, and get more capacity, performance, etc. , while the system administrators/system programmers update the underlying OS,…
Port away from? We moved everything TO POWER/AIX.
I've certainly seen my share of underutilized System p servers, but in the right hands, they're a powerful option, with features that Linux still hasn't caught up to yet. For a cool feature that I would love to see on Linux someday, peek at checkpoint-restart, which enables some really sweet capabilities [1]. Generally, anytime I hear a business tell me uptime reliability is nearly a blank-check requirement (when downtime losses are described in 5+ figures per minute and up, and it is a 24x7x365 operation), but they must still stay on a Unixen box because that is what their IT shop knows, and it's not a parallelizable application, I start looking in AIX's direction. It can be a right PITA to work with if you grew up with Linux or BSD, but once you figure out some add-ons to use, it gets much more pleasant to use.
[1] https://www.ibm.com/developerworks/aix/library/au-aix7.2.1-l...
Earlier quoted context omitted.
I could be wrong but I assume that there's enough inertia in the high-end business-to-business side of things (the part that matters for high-end server processors) that they probably can keep milking that cow for a few years more. People who have only been buying Intel CPU in the last decade won't jump ship immediately. Nobody gets fired for buying Intel CPUs, at least until very recently. You know you won't have co…
I suspect anyone ordering top-bin CPU skus by the pallet-load has an ongoing program for evaluating performance-normalized TCO of at least a few candidate platforms based around a few candidate processors.
Earlier quoted context omitted.
Newer steppings of Ryzen have a fix applied, and AMD is allowing people to RMA those processors affected. Threadripper is unaffected by the bug. Source: https://www.phoronix.com/scan.php?page=article&item=new-ryze...
AMD has never stated that there is a fix for this. If you look at the link you posted, this is Larabel making this claim on the basis that he hasn't seen any forum posts noting this problem on samples manufactured after week 25. Quote: > AMD has not provided an official public explanation of the fundamental problem, but from those in our forums and elsewhere, it appears to affect Ryzen CPUs manufactured prior to week…
What are you basing this on? I've seen no good reporting on this* but the AMD Community segfault thread (https://community.amd.com/message/2796982) seems to suggest that almost all Ryzen CPUs manufactured before the end of June (which I assume is most of the current stock) are affected. Very few people load their Ryzens sufficiently to hit this bug, but a large number of those who attempt it seem to get the segfault.
* I would like to know the percentage of processors with the bug, percentage of review units with the bug (I would be very interested in this), and maybe some long-term testing of "bug free" processors etc.
Earlier quoted context omitted.
Exactly, and for pretty much the same reason. Once CMOS prices got to the point where 3D graphics acceleration (not SGI's only core competency, but pretty close revenue-wise) was desired at consumer volume, they were toast. NVIDIA had they beat on almost everything that mattered by 1998 or so.
I remember when my scientician friends started switching from SGI to multiple super-beefed-up Dells in the early 2000's. They could get five monster custom Dell machines for the price of one Octane or Indigo or whatever. And they delivered better performance. (Or so they told me.)
Earlier quoted context omitted.
More than that, Intel builds custom chips to AWS specs.
Kind of. From what I understand, any customer large enough can get small features added, and they just get turned off for the other customers.
Q: Why is Intel charging so much? A: Because enough people pay. Q: Why do people pay so much? Why do they pay a high premium for top performance? A: Because, considering all factors, it's worth that much for them. For many it's too much but for many others it's ok. Q: But it's the same silicon! A: Manufacturing cost plays little role when the price is not dictated by a race to the bottom between different manufacture…
Now is the only time you can charge premium for new chips. Nobody will pay that in a year or two. In five years it's on Craigslist, and in ten, it's ancient crap you find on motherboards pulled from machines thrown into the dumpster.
Of course so does a $5 raspberry pi.
Earlier quoted context omitted.
Power is big iron reliable. if you can pay for it it likely pays off. Probably easier to manage a handful of million dollar Power/AIX setups than 1000s of servers.
I don't follow, this is the same POWER 'big iron' just running Linux instead of AIX.
Earlier quoted context omitted.
> Only Xeons offer ECC > And only [1] i5/i7 offer unlocked multiplers. These statements are false, which you yourself acknowledge later in your post. You don't get credit for noting a caveat when the caveat is that your point is entirely false, at least in my book. Intel no longer manufactures 2-core Xeons so Pentiums and i3s support ECC nowadays. This isn't uncommon, this has been the norm literally since they stopp…
There has been one Pentium and i3 that are unlocked since the start of the Nehalem era. What's up with the i3-6320 supporting ECC but the i3-7320 no longer supporting it? Intel is at the least, very inconsistent with ECC on the consumer side.