It is exactly a year ago when this blog started. Once intended as satire for an infamous blog site, I suppose today we can say that we have outperformed Scierikou in terms of accuracy and usefulness in predicting the future and providing analysis on current events. It is indeed facinating that what we have been discussing collectively for the past several months only ends up lately on someone else's blog. Throughout the year we've been accused of painting AMD in a bad light but we managed to prove that it's only because there's a thick dark cloud hanging over it. We've been brutal and honest but so is the truth.
I would just like to mention that this site never really kicked off until our friend Sharikou180 provided a link to this site. Site activity wouldn't have picked up if it wasn't for the people that come here and contribute. Many thanks everyone.
Looking ahead, I'm considering the option of opening up this blog for additional contributors. I can see that some put considerable effort in their post which are at times quite noteworthy and insightful enough of being an article by itself. The two options for contributing are adding authorised authors (requires blogger account), and as for our friends in the "industry" that requires anonymity, you can send your article through e-mail (I will post the article under your chosen pen-name).
What do you think?
12.03.2007
One!
Subscribe to:
Post Comments (Atom)
34 comments:
Hi, this is Chicagrafo.
Congratulations on your year. Your post reminded me of my second anniversary coming in January, and I wanted to ask you if you have an idea of how to "celebrate" it.
Regarding opening up for contributors, go for it, like you said, some are so good that deserve us to open the doors.
Hey roborat,
Nice articles overall. I only started looking at your blog very recently.
Keep it up.
Congrats on your first year. I for one appreciate a blog where you don't have to wonder either how your words will be twisted or whether they will just be buried. Keep up the great work.
This from scalability.org
The take home message is that a 2 GHz Opteron 2350 appears to seriously best a 3 GHz Xeon 5365 on a real test (about 22% faster in favor of the Barcelona). Moreover, the 2 GHz 2350 is about 6% slower than the 3 GHz Penryn in the worst case, and about the same in the best case.
He provides plots from the article (from Intel's web site no less) to back up his claim. The thing I noticed about the data is that it is SPECfp_rate and LS-DYNA data. Intel's previous weak point.
To me the data shows that Intel has made huge strides in closing the gap where AMD was hitting them hardest. And they have done this without quickpath (CSI). I'm really looking forward to seeing what Intel's architecture can do when Nehalem is released.
In fairness to the author he then goes on and points out that he is also excited by Intel's improvement integer performance.
I think he missed the boat though, because Intel's rate scores should only improve once the FSB is taken out of the equation.
Congrats on the success.
Happy 1st.
Jack
Congratulations Roborat! Here's to another very successful year :). Meanwhile, it seems that Scientia's blog is turning into a ghost town!
You guys thought the current Phenom performance is awful?!
It gets worse. It turns out that the reviews weren't done quite correctly:
http://techreport.com/discussions.x/13724
We don't yet have a BIOS with the workaround to test, but we've already discovered that our Phenom review overstates the performance of the 2.3GHz Phenom. We tested at a 2.3GHz core clock with a 2.0GHz north bridge clock, because AMD told us those speeds were representative of the Phenom 9600. Our production samples of the Phenom 9500 and 9600, however, have north bridge clocks of 1.8GHz. We've already confirmed lower scores in some benchmarks.
Given everything we've learned in the past few days, our review clearly overstates Phenom 9600 performance, as do (more likely than not) other reviews of the product. We can't know entirely by how much, though, until we can test a Phenom system with the TLB erratum workaround applied.
Things just keep getting worse for AMD. Meanwhile, I'll continue enjoying the blazing fast quad core speed I've enjoyed since April. I also fully intend on purchasing a Yorkfield quad core come January. This system will have had a nice long life. From a C2D E6600 to a Q6600 to a 45nm quad. I also hear that the Geforce 9 is due out in February. Depending on the performance I'll likely pick one of these up too!
Truth and accuracy can only be revealed and proved over time. That’s what this site is all about. That’s why I'm here, and it has been, indeed, a pleasure and an education.
KUDOS DOC!
SPARKS
GIANT! Great Link!
http://techreport.com/discussions.x/13724
This is a must read!
SPARKS
Congrats on one year, i found this site through 180Sharikou, I have always read scientia's blog but don't think i posted more than once or twice, mainly becasue he is so blind into thinking he is open minded, anyhow keep up the good work, i'll continue reading,
BTW my name came about becaue Penix posted some garbage (suprize suprize) and i had to prove him wrong)
giant
I also fully intend on purchasing a Yorkfield quad core come January.
I am thinking of skipping Penryn and directly going for Nehalem. That should be available in about 9 months. Any idea what SSE 4.2 includes on top of SSE 4?
I am thinking of building a V8 out of Nehalem. I hope Intel does not stumble on execution, now that AMD is no longer in the same league.
Congrats on building up this blog. As for guest posters/bloggers, email is a good option as some folks like myself (some call me Guru), prefer to remain anonymous.
Or perhaps you can just split off a freestanding blog (don't know if this is possible) for potential new topics or blog submissions - like having a blog suggestion/submssion link on the left of the main page.
Roborat- congratulations on your modest milestone, it's been enjoyable to read and participate here.
Guru- I think you can get a google/blogger account and still remain anonymous. Either way I do enjoy reading the commentary from you and the other people who are involved in CPU design and/or production, even if much of it sails over my head.
Congrats on your milestone. Sorry I haven't been able to contribute as much in the past year- the move to a startup has left no time for any sort of insightful commentary, and I ain't one just to make noise, unlike many on those "other" blogs.
I hope you and the others here continue to prop up the level of discussion in the blogosphere.
Happy Anniversary. Keep up the good work!
yeah, time has passed. I still remember why I started blogging (few months before C2D launched)... yeah because of stupid comments posted by Sharikou ALL over the tech related net, with his blog link advertised :).
And eventually Sci joined in the game, and at first I thought he would be much better than Sharikou but the reverse ... ;)
I enjoy reading/posting in this blog. GPTB! ;)
Dude, I love your blog. I haven't posted yet but I read your stuff all the time.
The analysis of facts is just awesome. You look at facts and come to a conclusion, rather than come to a conclusion and look for facts to support it.
You shouldn't be celebrating(jk), you should be looking for more information on this:
http://techreport.com/discussions.x/13087
and this:
http://techreport.com/blogs.x/swasson
It's just a single source, but I have a feeling we'll be getting confirmations of it soon enough.
Thank you Roborat.
www.amdzone.com
Turn up your speakers and listen CLOSELY, you can hear the crickets chirp.
The bad news on Barcelona/Pheromones has hit the Street fella’s, and it ain’t pretty. They took a 4.24 percent hit today, or 41 cents, this is big. This represent a low going back to July 2003.
Conversely, despite the overall market drop, INTC GAINED 6 cents on the dollar. If a market turn around comes into play, in the near future, plus a good 4Q earning report, INTC will take off.
Overall market conditions have been the limiting factor in INTC’s suppressed share price. INTC holding the line in this environment is no small feat. If the banks can get their act together…..!
SPARKS
http://moneycentral.msn.com/msn/
stock_quote?Symbol=amd
Then there is this. Can it get worse, I wonder?
http://www.hexus.net/content/
item.php?item=10493
YES, it can.
http://www.extremetech.com/
article2/0,1697,2226939,00.asp
SPARKS
Thanks giant, I dug about the things you mention in your comment and wrote a blog about it:
more-terrible-news-about-phenom-and-k10.html
The bad news on Barcelona/Pheromones has hit the Street fella’s, and it ain’t pretty. They took a 4.24 percent hit today, or 41 cents, this is big. This represent a low going back to July 2003.
Indeed, at the same time AMD's value dropped to 5.1bn. That's LESS than what they paid for Ati a little over a year ago!
I am thinking of skipping Penryn and directly going for Nehalem. That should be available in about 9 months. Any idea what SSE 4.2 includes on top of SSE 4?
I am thinking of building a V8 out of Nehalem. I hope Intel does not stumble on execution, now that AMD is no longer in the same league.
I'll want to see the retail reviews on the new Yorkfield quad cores. I'm especially interested in the overclocking.
As for Nehalem, it's going to be a monster. It's going to be crazy. I can't wait to get my hands on an OCTA core Nehalem CPU! :D My Houdini rendering will just fly on that sort of a setup.
Pat Gelsinger describes some of the Nehalem specific SSE4 instructions in his IDF keynote. AFAIK, Nehalem quad core is due out in H2'08, while Nehalem octa-core is due out in 2009.
I predicted this months ago, and it has happened. AMD is making No friends in the press.
Read this as HEXUS rakes WRECTUM RUINZ over the coals.
http://www.hexus.net/content/item.php?item=10585
SPARKS
I'll want to see the retail reviews on the new Yorkfield quad cores. I'm especially interested in the overclocking
The eight-core Nehalem won't be cheap. And in DP configuration, it will have 32 threads--even Vista won't scale well to those many threads.
core2dude
"And in DP configuration, it will have 32 threads--even Vista won't scale well to those many threads."
Who uses Windows with hihg-peroformance CPUs anyways. Linux all the way ;)
I'll likely get a single socket Nehalem system. Quad or Octa core would be excellent. That's eight to sixteen threads.
Apps like Houdini scale well with threads anyway. That, coupled with Nehalem's other performance enhancements will make it perform far faster than anything before it in terms of x86 CPUs.
Hmmm... Q9300 (45nm quad, 2.5GHz, though only 6MB total L2 cache) will go for ~$270 during the 45nm launch Jan 20. Oh I guess I should say this is an Intel chip, if the 45nm or 2.5GHz didn't give it away!
And so the upperbound for all near term AMD quads.
And now the question is, what is better for AMD from a biz perspective a $200 quad core or 2, $100-120 dual core K8's (I think the K8's would probably be better for profit). I guess it is more important for Hector to save face than make money?
AMD is at $9.09 and falling. INTC is holding steady at $27 - or nearly triple AMD's stock price.
AMD is continuing with K8 on the desktop:
http://www.dailytech.com/AMD+Resurrects+K8+Architecture+for+2008+Roadmap/article9899.htm
They need some product to sell for 2008, since Phenom was shipped with serious bugs. To fix these bugs causes 20% performance penalty, dropping K10 below K8 performance.
Yuck giant, don't compare stock prices like that. You have to compare them in terms of gain/loss PERCENTAGE over a period of time.
Giant - nice link. This is further supporting evidence of the 65nm disaster.
They are moving over all the 90nm products to 65nm, but cutting the top 2 bins? Hmmmm... let's take our highest price part and not produce it on 65nm - what other reason could there be other than they CAN'T produce it on 65nm?!?
Well now that I think about it perhaps they don't want to go through the embarassment of having their K8 65nm beat or match K10?!?
Also notice they are cutting L2 cache - some may say to make dies size smaller, but this should also help the thermals out a bit too....
At least it does look they have finally swallowed their pride and realized it is more profitable short term to make K8's until they get K10 sorted out. Unfortunately for their shareholders, the market had to "hammer" (pun intended) the AMD stock price for Ruiz to finally remember he is running a business. There are times where he looks like he is just winging it though and lacks a short and long term plan (other than 'profitability is a priority'. Gee, thanks for that nugget!)
"Also notice they are cutting L2 cache - some may say to make dies size smaller, but this should also help the thermals out a bit too...."
This also exposes another AMDism .... when Brisbane launched with higher L2 latencies, they said the wanted to accomodate larger caches in the future... but it appears they are not going to, nor had plans to, increase that cache.
Truth is... they L2 increased latency is yet a further indicator that the 65 nm process is a bit lacking.
It is truly a nice and helpful piece of information. I am satisfied that
you just shared this useful information with us. Please
stay us up to date like this. Thank you for sharing.
Feel free to surf to my web site weight loss calculator
Post a Comment