Not to spoil you and Caamsa's party, but the survey also included those Joe Sixpacks, who buys their computer from Dell and HP. So of course due to Intel's market share, and their ability to actually deliver in volume, more people use them.
In a overclocking community though, the survey does not apply.
P.S. Sharidouche once used this report to "back up" his claim that "Intel's engineers are also dumber than AMD's". Someone needs to recommend a good psychiatrist to him.
But i sell computers in a computer store, and you don't want to see faces of consumers who buys intel based computers. They really look dumb.
he, he, just kidding
---------------
"Contrariwise," continued Tweedledee, "if it was so, it might be; and if it were so, it would be; but as it isn't, it ain't. That's logic."
But i sell computers in a computer store, and you don't want to see faces of consumers who buys intel based computers. They really look dumb.
he, he, just kidding
Yeh I know what you mean... I used to work at Fry's, and you don't want to know how... computer "illiterate" people are when they just come in and ask for a computer.
"Dual core? So does that mean I 2x the clockspeed?"
Because an unlocked multiplier isn't going to do much for you when the chip itself has very poor overclocking potential. In fact, an unlocked multiplier on a Phenom is redundant. Thus, they're making you think you're getting something special, when in actuality you're not.
Gosh just like buying one of those Green versions of the segate and western digital drives.....hmmm you get the same performance but it uses less power and eco friendly storage....hmmm a green box...................must be some kind of gimick.......................like those dang black box edition thingys.
Oh. So I would assume all the people overclocking never change core voltage.
WAIT...
Upping the core voltage IS PART OF OVERCLOCKING. So if you can't up the core voltage you can't overclock.
What was AMD trying to hide by not giving hardware to reviewers and instead controlling the environment and placing restrictions and NDA's on the press?
I think we all know now, we've seen the benchmarks, and we've read about Phenoms being recalled even after they were on the trucks to the retailers for launch.
Hats off to AMD for doing the right thing and recalling the processors, but a big thumbs down for being a failure of a company that hasn't figured out how to launch a product for over a year and a half now.
You need to raise the voltage in order to over clock? Hmmm my super socket 7 board did not have that fancy option and I had to change a jumper to raise the FSB.
So the medicaition is not helping is it.................
You need to raise the voltage in order to over clock? Hmmm my super socket 7 board did not have that fancy option and I had to change a jumper to raise the FSB.
So the medicaition is not helping is it.................
Well, since you have overclocked a processor this decade I recommend that you do some reading in the overclocking forum.
Well, since you have overclocked a processor this decade I recommend that you do some reading in the overclocking forum.
I overclocked a tutalan processor once just buy raising the FSB on a SOYO mother board. It had no option to raise the voltage. I was able to take it from 1.4 to 1.6 and it was rock solid. So you don't aways have to do the voltage thingy...........
Same with my Barton 2500+ raised the FSB from 133 to 166 and it went from a 2500+ to a 3200+ with no change in the voltage on a MSI Nforce2 mb.
O yea and I forgot, I had a computer once.....with this button......a turbo button....and when I pushed the button the computer ran faster and I was able to Wolfenstein 3D.............
This spring I am going to over clock my clock by one whole hour and not even increase the voltage one bit.
Message edited by caamsa on 11-29-2007 at 08:09:08 PM
Chips could have different cache size levels, different numbers of transistors, different built-in features, numbers of cores, etc. Thus a person needing only a basic chip could pick out a simple cheap chip, like the Celeron, for simple tasks, while someone needing a chip for complex business needs could pick out a different one, and a gamer could pick out another chip altogether. Even among these three styles of chips, there could be a number of varieties based on real mechanical differences, rather than than just a multiplier. A benefit to the company would be that it would spend less money programing the chips and less money in advertizing, while the consumer could benefit by having an easier, more clear cut choice among chips.
And this greatly increases costs.
Again, Price goes up.
The cost to design and maintain so many different chips is much greater than setting a default value.
Major parts of the entire economy is based on this.
Why do companies charge more for 10,000 Licensed copies of software than a single copy when all a company gets is a single DVD and a license code?
Why Am I not allowed to make copies of a DVD I purchase and resell at a lower cost than I bought my one copy?
The whole point is that each company is driven to make as much profit as possible.
Profits are what drives the economy.
Without companies striving for and making profits, we would all be in worse shape.
And this greatly increases costs.
Again, Price goes up.
The cost to design and maintain so many different chips is much greater than setting a default value.
The point of a modular design is to go around this at the die level. Graphics cards are already made this way; that's because it is easy to increase or decrease the number of execution units when most of your chip runs parallel pipelined code.
The business question is whether the same concept can be applied to CPU dies, especially as they get more complex. Are the die space savings (e.g, omitting a multimedia unit from a server CPU) worth the R&D and production expenses (designing modules to tighter tolerances, making new masks, adapting production to new masks, die space overhead).
It seems the heads at AMD and Intel both predicted yes, with the way they characterized Fusion and Nehalem, respectively, on the roadmap.
The other analogies you gave aren't relevant because the cost savings here don't require a full redesign-retest cycle.
Yeh I know what you mean... I used to work at Fry's, and you don't want to know how... computer "illiterate" people are when they just come in and ask for a computer.
"Dual core? So does that mean I 2x the clockspeed?"
Then you understand me. Town where i live has less citizen then your street. So, my point was to say that only "redneck" buy already builded computers. As you all knows, we (in Croatia) have industrial "delay" over america for about 5 years. So when somebody comes to store where i work, and says that he wants intel because AMD is warmer, you know why i said that you don't want to see their faces.
Yomamafor, i didn't have a good opinion about you, until now. You always insult BM, and i didn't like that, but now i know you are one "smart MF".
Restepk (ALI G)
Then you understand me. Town where i live has less citizen then your street. So, my point was to say that only "redneck" buy already builded computers. As you all knows, we (in Croatia) have industrial "delay" over america for about 5 years. So when somebody comes to store where i work, and says that he wants intel because AMD is warmer, you know why i said that you don't want to see their faces.
Yomamafor, i didn't have a good opinion about you, until now. You always insult BM, and i didn't like that, but now i know you are one "smart MF".
Restepk (ALI G)
Hey, I understand some people completely underestimated AMD's capability, and the quality and ingenuity in their products. I also feel infuriated when someone who regards AMD as "3rd world country, cheap knock off manufacture".
Quote :
If your son has requested a new "processor" from a company called "AMD", this is genuine cause for alarm. AMD is a third-world based company who make inferior, "knock-off" copies of American processor chips. They use child labor extensively in their third world sweatshops, and they deliberately disable the security features that American processor makers, such as Intel, use to prevent hacking. AMD chips are never sold in stores, and you will most likely be told that you have to order them from internet sites. Do not buy this chip! This is one request that you must refuse your son, if you are to have any hope of raising him well.
While this was supposed to be a joke, some people actually took that comment to heart, and argue for it as if they know AMD inside out.
As for BM, I attack him because of his inability to admit his constant spewing of FUD. If you look closely to his comments, you'll realize that they're all just sugar-coated nonsense. When confronted with real data and factual information, he readily disregard them as "rabid reactions", and starts his childish personal attacks. He has been doing this for the past 2 years, yet no one called him out for it.
You can look for his "Barcelona is 40% faster in various workload" argument with me, then you'll know exactly what I mean.
Seriously, the drive to lock the multipliers began in the early 90s when people were remarking Intel parts (yes, in some cases, even filing down the laser-marked ceramic on the original Pentiums) and they'd fail at their "new" speed. Intel would get the customer return when it didn't work and have to tell the customer they got screwed by a vendor chain with some unscrupulous behavior in it. It was a bad deal all around.
Intel is not opposed to overclocking as a matter of policy-- they have experimented with technologies which can (and do) lock the allowable FSB, and discarded them because widespread overclocking is an enthusiast matter and there's just no reason to alienate the enthusiasts in this way.
As for designing lots of little "microchanges" into an architecture for mechanically different dice to distinguish pricewise as Sailer suggests, that's way too much cost and effort to be worthwhile. Too much validation involved in dice with several different featuresets vs just several different speeds. Removing half a cache can be done efficiently as you can design it on the edge of the die and essentially saw it off during singulation; having a feature on die and deactivating and selling it as a lower price is primarily a pure yield-savings measure-- sell something which doesn't fully work (as long as there's a market for it) instead of throwing it out entirely (see AMDs 3-core chips). Much easier to rate the products by speed and some minimal number of features (number of cores, amount of cache, onboard graphics or not) because otherwise your validation costs and time-to-market go straight to hell.
wr is correct in that more modular designs are in the making, but I assure you each "modular" design will have... several different speeds for different price points. So, yes, mucky-mucks at Intel and AMD may have chosen to have more designs out there in the field with some degree of modularity to the design, but at the same time they've accepted the added validation overhead. No, it may not be a full validation cycle, but there's a significant resource requirement for each modular design. You change a die design-- even a minor one-- you change possible speed paths, and that takes time to shake out. You can accept or mitigate the risks with modular design, but that doesn't make them (or their costs) go away entirely.
Message edited by archibael on 11-30-2007 at 02:06:16 AM
In the days of unlocked CPUs, some dogdy retailers used to raise the multiplier and sell the CPU at higher pricepoints. Hence the locking of the multis.
Or so they say anyway... it was a long time ago.
Well, we live in the Internet age now... if one retailer did something like that it'd be all over the internet in a matter of minutes... so hopefully no manufacturer is still locking multipliers under the guise of consumer protection.
Hey, I understand some people completely underestimated AMD's capability, and the quality and ingenuity in their products. I also feel infuriated when someone who regards AMD as "3rd world country, cheap knock off manufacture".
Quote :
If your son has requested a new "processor" from a company called "AMD", this is genuine cause for alarm. AMD is a third-world based company who make inferior, "knock-off" copies of American processor chips. They use child labor extensively in their third world sweatshops, and they deliberately disable the security features that American processor makers, such as Intel, use to prevent hacking. AMD chips are never sold in stores, and you will most likely be told that you have to order them from internet sites. Do not buy this chip! This is one request that you must refuse your son, if you are to have any hope of raising him well.
While this was supposed to be a joke, some people actually took that comment to heart, and argue for it as if they know AMD inside out.
As for BM, I attack him because of his inability to admit his constant spewing of FUD. If you look closely to his comments, you'll realize that they're all just sugar-coated nonsense. When confronted with real data and factual information, he readily disregard them as "rabid reactions", and starts his childish personal attacks. He has been doing this for the past 2 years, yet no one called him out for it.
You can look for his "Barcelona is 40% faster in various workload" argument with me, then you'll know exactly what I mean.
I would not say that no one calls him out on it. A lot of people call him out on it. Also I would not say that he lashes out in a childish way anymore than a lot of people do in these forums.
I think he truly believes what he posts to be true based upon his over optimistic view of AMD and hatred of Intel. I don't think he is coming up with these numbers on his own it is stuff that he has probably read and then posting. How would he know benchmarks etc. ..... unless he reads them some where.
You should go and do some searching in the forms regarding BM and you will find plenty of "people calling him out" as you say.
I would not say that no one calls him out on it. A lot of people call him out on it. Also I would not say that he lashes out in a childish way anymore than a lot of people do in these forums.
Actually, BM started doing this 2 years ago, and it has never stopped. Can you say the same for the rest of the forum?
Quote :
I think he truly believes what he posts to be true based upon his over optimistic view of AMD and hatred of Intel.
If he truly believes what he posts, then he should be able to back it up with solid facts, even its technical data. Yet he presented none. When confronted with solid facts, instead of presenting his own data, he used editorials to prove his points. When he was asked to provide links, he provided a blog. How is that solid evidences? How is that real technical data?
People want to be optimistic about AMD, but not overly-optimistic that its not in touch with reality anymore. Back in the Core 2 days, he claimed C2 will suffer from cache thrashing. It was proven wrong. Then he claimed AM2 would perform on par with Core 2, which was also proven wrong. Then he said 4x4 will be the ultimate enthusiast platform, which turned out, performed lower than lowered clocked Q6700 by a wide margin. Then he claimed K10 will annihilate any Core 2 at 1Ghz; do you think he's right? This is not optimism. This is simply his "opinions", based on his preference for AMD, with no solid facts. Yet he claimed it as if its the truth. To be honest, he's just like the boy who cried wolf. If he's done frequent enough, what do you think his credibility will be?
Quote :
I don't think he is coming up with these numbers on his own it is stuff that he has probably read and then posting. How would he know benchmarks etc. ..... unless he reads them some where.
Then why does he not present it when asked? We're in a modern world, where information is readily available on the internet.
There was a respectable forum back in the summer, with the name Jack. He also made a lot of claims, but he not only backed them up with actual technical information and data, he also carefully explained them so everyone knows what he's talking about. He gained a lot of respect from us, because he basically shared his knowledge. He once claimed AMD's 65nm is suffering problem, which is evident now. He also claimed Intel's HK/MG to be one of the innovative technologies in the industry, which appears to be true.
On the other hand, BM also made a lot of claims, but almost none of them is backed up with the right technical information, or even links. He would just post something like, "I believe blah blah blah will give blah blah and increase in performance about xx%". When confronted, he basically said, "you guys are all AMD haters". How is "I want you to provide more information in regard to your claim" related to "I am an AMD hater"?
Quote :
You should go and do some searching in the forms regarding BM and you will find plenty of "people calling him out" as you say.
Then you should know better to think twice before believing his claims.
