I agree. Only the EE version will be a lot cooler in comparison to both non-EE Athlon 64 and desktop Athlon XP.
Actually, just to throw a bit of fuel on the fire of #3, there was a time when this was true! No, I'm not referring to what is known today as OS disk formatting (letting the operating system lay down its patterns as per the article). I'm talking about the REALLY old days, when IDE drives were just appearing on the scene with a massive 130mb of storage. The previous technology, MFM and RLL2,7 hard drives, required what was called "low-level formatting", done with commands given directly through the drive controller card BIOS, because the disk drive hardware itself was stone-stupid. The controller was given the number of cylinders, sectors and heads available on the drive, it then wrote the sector information onto the media. A drive formatted with a Western Digital MFM controller was unable to be run by an Adaptec controller, simply because the servo information written on the disk was incompatible. Early IDE implementations had a flaw, in that you could command them to try the "low-level format" operation on an IDE drive. More than a few times, the drive would attempt it and blow the factory servo track formatting. You had to send the drive back to the source manufacturer to have it reformatted. Improvements to the drive firmware slammed that door shut, and you have the near bulletproof IDE/SATA drives of today. #3 is virtually impossible to do, but not totally. As one of the Murphy's Law correlaries states, "Nothing can be made foolproof, because fools are so ingenious!". Thus endeth the PC Ancient History lesson for today.
Myth #2 source? I have a possible source for the Myth number 2 concerning "too much space corrupting your hard drive". This myth could have origins int he Unix world with file server quotas when mounting directories under NFS mounts. Not all flavors of unix/Linux are created equal, and NFS can be one of the Unix administrator's biggest challenges. Not all systems handle NSF mounts and quota notifications equally either. I have whitnessed users' environments completely die and their e mail go completely bye-bye because the mail application or another application's log files didn't know how to interperate the quota limit. Of course this DOES NOT apply to a local files system (hard drive) but only applies when using NFS. In this case the user data could get lost when exceeding a quota. Likewise, if the base volume on some file servers fill up, the file server can crash hard. Again, NOT the hard drive on a local work station. These are special cases.
Very Sad AMD-Intel Myths over here are almost as known as chapters in the Holy Scriptures. "Is there a way to exorcize Intel´s marketing lies?" Somebody knew oracle8.x cannot run on P4 because a cpu id bug? Does it means Athlon XP was more stable for oracle? <-- not true, but hey, let me lie as intel always does. Cant we create a myth like, intel is coupled with pepsicola to take over the world so more people buy amd?
thinking back to those days, I had a friend telling me about that too. a 40MB disk. Seems that low level formating from the bios tends to yield different capacities each time. So ended up formating it like 5 times in a day, trying his luck to get a larger capacity, then it died on him on the 5th time.
One thing I might have missed but was looking for, so if I'm repeating something here forgive me... but it seems like nothign was mentuioned in the AMD vs Intel stability myth about the chipsets. It has always seemed to me that you can get an equally unstable platform out of both CPU's by cheaping out on the motherboard and running poorly designed chipsets, even at proper operating specifications. It seems like intel has always done well to provide high performance chipsets for the intel platform, whereas AMD typically provides chipsets geared at their high end server platform, relying on other manufacturers to provide the beefy stuff for high performance gaming and what not. Of course this changes witht he ATI merger, and I am really curious to see if we see a shift in the myth over the next 5 years as a result.
IIRC, all Float operations except a few notable non compliant instructions sets follow the IEEE 754-1988 or 200X Standard. The S E F or Sign, Exponent and Fraction/Mantissa should be the same when the standard is rigidly conformed to. Although i do recall Pentiums having a problem with a 1 in a billion calculation round off inaccuracy due to bugs in the mircocode. Don't know if that bug is dead lol, maybe someone should look up the Intel Release sheets. or maybe the precision mode in use is different? I'm not too sure about it. But if i recall correctly, all Architectures that follow the IEEE 754 should stick with a precise computational method and set of rules, only logic isn't governed. Or it might be input error although i doubt TCP's CRC wouldn't catch it. Cheers.
The guide has just been updated! Join us today as we take a look at the most common and possibly the most stupid tech myths in the world, and BUST them! Here is the update:- Link : Tech Myths B.U.S.T.E.D. Rev. 1.2!
I think a better PSU myth to bust is how much power a PC really uses. The power supply calculators that are around the web are making people think that they need a 500W power supply. Like you mentioned, a crappy 500W power supply probably can't even deliver 300W. But the thing is, most people don't even need 300W. If you're not overclocking like crazy and only have a single video card, a good 300W will handle you no-problem. Your first PSU myth isn't helping by suggesting you buy the biggest PSU you can as it won't be using any more electricity than a lower powered model. But a good 300W PSU will run you about $50 vs. $100+ for a good 500W PSU. And I would rather keep the $50. And in some ways, 300W PSUs will use less power than 500W PSUs as the efficiency of a PSU is not linear. They tend to peak on the mid-high end of their range and drop off on the low end. With low-powered Core2 and AMD CPUs these days, it not unusual for a low-mid power computer to draw between 80-150W max. A 300W or less PSU will most likely be more efficient at this load and will draw less AC power than a 500W PSU. Silent PC Review has a great article about PSUs titled Power Supply Fundamentals & Recommendations. This article has some good myth-busting info about power supplies.
You raised some really interesting points which are not covered by article. But what you said is not completely true. A quick glance over some of the PSU reviews at SPCR, most PSUs peak at mid range, NOT the high end range. So a 500W PSU is most likely more efficient at 250W thereabout. Most PSU's efficiency will only drop off below the 20% of the maximum rating of a PSU. And it's possible that a 500W PSU is still more efficient than 300W at 80-150W usage range.
I suppose it's quite difficult to make blanket statements about this kind of thing, especially with the huge variation in PSUs on the market
I am wondering about #5. The reason is because I keep on hearing about AMD based servers having problems, and Intel ones being fine. I read the explanation, but I am confused because so many people I have talked to say otherwise. Would someone like to explain?
i like the TechARP Mythbuster series. how about talking about LCD response time? people have been religiously saying that when you're a gamer, you need low response time (8ms?) but as a gamer myself my 151s with 30ms response time isn't giving me ghosting effect and it works great on all the games i play.
Myth? I don't think so. There's definitely ghosting, but whether it annoys you or not, that really depends on individual. For me, the 2001FP is pretty good, no obvious ghosting.
Subjective to a certain extent, yes I agree with that too. But the real issue is how most people seem to blindly go along with the response time spec when in reality it's mostly bs
yeah.. like 2ms response.. the retailers @lowyat go WHOAAAAAAAAAAAAAAAAAAAAAAAAAAaaAAAAAAaa.. "2ms response time ar..i tell you ar, your game long chong play sui sui ar.."
