lane@cs.dal.ca (John Wright/Dr. Pat Lane) (03/06/90)
I wonder why computer mags and people here on the net spend so
much time comparing various clone computer brands, most of them from
from mail-order houses in the U.S. Surely all these brands are assembled
from the same generic components that you can buy from these places.
Why not review/discuss the components, in particular the motherboards
and then, when reviewing the "Brand X" PC, simply say "well, this machine
has a 'Kung-Foo 286' motherboard (ficticious names here, of course!) -
see issue XXX, p. nnn for a review of that" and list the various
component that the manufacturer/assembler has used.
Around these parts people generally choose from a brand-name machine (IBM,
Compaq, Zenith, etc.) from a store like Computerland, or, a no-name clone
from one of the smaller stores. Many of the clones are assembled in a
back room of the store! Others are assembled by mail-order houses in
Upper Canada. Most don't even bother to put a name-plate anywhere on
the outside. They all seem to be built from the same components so why
not choose your system based on the components.
I'm never going to see a review of "Joe-down-the-street's clone" in PC
Magazine and I'm not going to by from CompuAdd when I can get virtually
the same thing locally so most of those reviews are pretty useless to me.
I suppose the obvious answer is that PC Magazine is giving free publicity
to the same large U.S. mail order houses that advertise in the magazine.
Oh well, just rambling.
--
John Wright ////////////////// Phone: 902-424-3805 or 902-424-6527
Post: c/o Dr Pat Lane, Biology Dept, Dalhousie U, Halifax N.S., CANADA B3H-4H8
Cdn/Eannet:lane@cs.dal.cdn Uucp:lane@dalcs.uucp or {uunet watmath}!dalcs!lane
Arpa:lane%dalcs.uucp@uunet.uu.net Internet:lane@cs.dal.catims@starfish.Convergent.COM (Tim Simmons) (03/09/90)
lane@cs.dal.ca (John Wright/Dr. Pat Lane) writes: >I wonder why computer mags and people here on the net spend so >much time comparing various clone computer brands, most of them from >from mail-order houses in the U.S. Surely all these brands are assembled >from the same generic components that you can buy from these places. >Why not review/discuss the components, in particular the motherboards >and then, when reviewing the "Brand X" PC, simply say "well, this machine >has a 'Kung-Foo 286' motherboard (ficticious names here, of course!) - >see issue XXX, p. nnn for a review of that" and list the various >component that the manufacturer/assembler has used. >I'm never going to see a review of "Joe-down-the-street's clone" in PC >Magazine and I'm not going to by from CompuAdd when I can get virtually >the same thing locally so most of those reviews are pretty useless to me. >I suppose the obvious answer is that PC Magazine is giving free publicity >to the same large U.S. mail order houses that advertise in the magazine. I agree with this observation 110%. Have have gone through this same line of logic with individuals in the market for IBM clones. They ask "Should I buy a Northgate or a Dell or from Tran Nguyen down the street?" It surprises me how many people (many of them in the computer industry) don't understand what a clone really is. In most cases the only difference is the name on the front of the case. People have this idea that the company who's name is on thefront of the box actually manufactures the stuff that is inside the box. And they also think that company X's box is somehow different and better/worse than comany Z's box. This is how it works: A guy starts a clone company, he has some stickers with his company name printed on them, some business cards and some letterhead. He gets a resale license goes to the local PC parts distributor, or the parts manufacturer (Teac, Segate, WD, NEC etc) and buys his parts. motherboards, floppies, hard drives, VGA cards, case, keyboard and so one. He slaps all the parts together and puts his sticker on the front of the box. Meantime the guy down the street does the exact same thing, same distrubutors/manufacturers. Then the customer gets all confused and asks which one should I buy? Of course you have the big name clones like Dell who mold their own plastics so that thier cases look different from MR. smalltime PC assembler on the corner. Then the customer thinks "Wow they are a big company and thier box looks really cool, so I guess thier system must be worth $500 more than Mr. small time PC assembler" The point to all this is, the name on the box doesn't make a rats ass worth of difference. What is important is the components that Mr. small time is using, what service is offered and the price. DON'T BUY A PC JUST BECAUSE THE NAME IS FAMILIAR. >-- >John Wright ////////////////// Phone: 902-424-3805 or 902-424-6527 >Post: c/o Dr Pat Lane, Biology Dept, Dalhousie U, Halifax N.S., CANADA B3H-4H8 >Cdn/Eannet:lane@cs.dal.cdn Uucp:lane@dalcs.uucp or {uunet watmath}!dalcs!lane >Arpa:lane%dalcs.uucp@uunet.uu.net Internet:lane@cs.dal.ca
ilan343@violet.berkeley.edu (03/09/90)
>lane@cs.dal.ca (John Wright/Dr. Pat Lane) writes: > > >>I suppose the obvious answer is that PC Magazine is giving free publicity >>to the same large U.S. mail order houses that advertise in the magazine. > In fairness to PC-Mag and the likes, this doesn't seem to be the case. Last month I was shopping for a 386 clone. I went to clone-maker in So. San Francisco which had their machine assigned the Editor's Choice in a PC Mag round up of 20Mz 386's. This company doesn't advertize in any of the glossy pc magazines. As a matter of fact, they send machines for review as form of exposure. Assuming that PC magazine returns the test machines, they don't derive any direct benefit from the positive review. I agree with the main thread of the original posting. Most of the parts in these clones are generic components. Other than price and support, there is very little difference between "Joe's down the corner" and the higher-volume mail-order houses.
gt0159a@prism.gatech.EDU (LEVINSON,MARC LOUIS) (03/09/90)
>In most cases the only difference is the name on the front of the >case. >People have this idea that the company who's name is on thefront of >the box actually manufactures the stuff that is inside the box. How do you think IBM made their PC in the first place? All they did was assemble components, most of which were off the shelf items! >He slaps all the parts together and puts his sticker on the front >of the box. Meantime the guy down the street does the exact same >thing, same distrubutors/manufacturers. >Then the customer gets all confused and asks which one should I >buy? Some companies work very hard to make certain that the components they "slap together" work properly as a complete unit and yield top performance and compatibility. Many of the big name clone companies actiually build or spec their own designs now, although they may have used another's when they started. > >The point to all this is, the name on the box doesn't make a rats >ass worth of difference. >What is important is the components that Mr. small time is using, >what service is offered and the price. Buying on price is a good way to get burned! > >DON'T BUY A PC JUST BECAUSE THE NAME IS FAMILIAR. Agreed. Components, performance, compatibility are important. Most important, however, is to buy a box which is backed by a lasting commitment and warranty. Many of the more reputable clones include or offer onsite warranty repair by a nationwide service organization such as TRW. Believe me, it is worth it! I find that PC Magazine's reviews give a certain credibility to the vendor/clonemaker's claims, or disprove them. No I don't take their word for the absolute truth, but I do use it as one of the measures of comparison. PLEASE NOTE: PC Magazine makes every effort to list the exact components used in the box, therefore, tey review the parts as well. ------------------------------------------------------------------------- The opinions expressed above are my own, please note that they are OPINIONS, not absolute truths, although they do have some basis in fact. -Marc --------------------------------------------------------------------------- >>-- >>John Wright ////////////////// Phone: 902-424-3805 or 902-424-6527 >>Post: c/o Dr Pat Lane, Biology Dept, Dalhousie U, Halifax N.S., CANADA B3H-4H8 >>Cdn/Eannet:lane@cs.dal.cdn Uucp:lane@dalcs.uucp or {uunet watmath}!dalcs!lane >>Arpa:lane%dalcs.uucp@uunet.uu.net Internet:lane@cs.dal.ca -- LEVINSON,MARC LOUIS Georgia Institute of Technology, Atlanta Georgia, 30332 uucp: ...!{allegra,amd,hplabs,seismo,ut-ngp}!gatech!prism!gt0159a ARPA: gt0159a@prism.gatech.edu or marc@isye.gatech.edu
stecz@hotwheel.dell.com (John Steczkowski) (03/09/90)
In article <1592@starfish.Convergent.COM>, tims@starfish.Convergent.COM (Tim Simmons) writes: > > Of course you have the big name clones like Dell who mold their own > plastics so that thier cases look different from MR. smalltime PC > assembler on the corner. Then the customer thinks "Wow they are a > big company and thier box looks really cool, so I guess thier > system must be worth $500 more than Mr. small time PC assembler" Dell's Motherboard are of our own design and manufactured exclusively for Dell PCs. It is true that our peripherals are OEMed, but IBM even OEMs a lot of their peripherals. > >John Wright ////////////////// Phone: 902-424-3805 or 902-424-6527 John Steczkowski Dell Unix Customer Support P#: (512) 343-3571 9505 Arboretum Blvd. !s: uunet!dell!hotwheel!stecz Austin, Texas 78759 @s: stecz@hotwheel.dell.com x
dlow@hpspcoi.HP.COM (Danny Low) (03/10/90)
>In most cases the only difference is the name on the front of the >case. >And they also think that company X's box is somehow different and >better/worse than comany Z's box. >This is how it works: This is your basic incomplete and dangerous information. The description of how it works is a generality that has many many exceptions. The larger clone makers are certainly exceptions to this rule. They have qualification programs. This means the clone maker only buy from suppliers who can meet certain standards of quality. This means testing components to insure that the standards are actually being made. The result is you are less likely to get a clone that is DOA. Returning a mail order clone that is DOA is annoying and time consuming. From the MANY conflicting "I have no problems with my X clone" and "I will never buy an X clone again" postings on the net, there are some clone makers who clearly have quality control problems. The larger clone makers have their own IC design teams to design customer circuits for their clones. The resulting clone may be faster, cheaper or more reliable than a clone made entirely of off-the-shelf components. Check the motherboards on the clone. The small clone makers buy their motherboards off the shelf from an oem supplier. The large clone makers make their own motherboards. A custom motherboard is a good sign that this is not just another clone. I work for a clone maker testing new clones. Most of what goes in our Vectras are custom designed by us. The stuff we oem off the shelf has to meet our internal Strife standards. InfoWorld wrote an article recently on the tests an oem component has to pass to qualify before we will use it. You can bet a clone maker operating out of a garage does not do this sort of testing. You may be getting an equally good product (after all that is how Hewlett-Packard got started) but that is unlikely. Danny Low Disclaimer:These views are my own and not those of my company. Hewlett-Packard Sunnyvale Personal Computer Division HP SPCD dlow%hpspcoi@hplabs.hp.com ...!hplabs!hpspcoi!dlow
marshall@wind55.seri.gov (Marshall L. Buhl) (03/10/90)
stecz@hotwheel.dell.com (John Steczkowski) writes: >In article <1592@starfish.Convergent.COM>, tims@starfish.Convergent.COM >(Tim Simmons) writes: >> >> Of course you have the big name clones like Dell who mold their own >> plastics so that thier cases look different from MR. smalltime PC >> assembler on the corner. Then the customer thinks "Wow they are a >> big company and thier box looks really cool, so I guess thier >> system must be worth $500 more than Mr. small time PC assembler" >Dell's Motherboard are of our own design and manufactured exclusively for >Dell PCs. It is true that our peripherals are OEMed, but IBM even OEMs >a lot of their peripherals. Not only that, but Dell will probably still be around when you finally send your computer off to the great bit-bucket in the sky. A lot of smaller companies will fail in the next few years. I'm willing to pay extra for the comfort of dealing with a stable company. -- Marshall L. Buhl, Jr. EMAIL: marshall@wind55.seri.gov Senior Computer Engineer VOICE: (303)231-1014 Wind Research Branch 1617 Cole Blvd., Golden, CO 80401-3393 Solar Energy Research Institute Solar - safe energy for a healthy future
karl@ddsw1.MCS.COM (Karl Denninger) (03/11/90)
Time to shoot down some more misconceptions... >----- >Response 4 of 4 (9561) by dlow at hpspcoi.HP.COM on Sat 10 Mar 90 03:43 >[Danny Low] >(40 lines) > >>In most cases the only difference is the name on the front of the >>case. >>And they also think that company X's box is somehow different and >>better/worse than comany Z's box. >>This is how it works: > >This is your basic incomplete and dangerous information. The description >of how it works is a generality that has many many exceptions. The larger >clone makers are certainly exceptions to this rule. > >They have qualification programs. This means the clone maker only >buy from suppliers who can meet certain standards of quality. This means >testing components to insure that the standards are actually being >made. The result is you are less likely to get a clone that is DOA. >Returning a mail order clone that is DOA is annoying and time consuming. >From the MANY conflicting "I have no problems with my X clone" and "I >will never buy an X clone again" postings on the net, there are some >clone makers who clearly have quality control problems. Yep, including some of those who do their testing. Then there are those with an excellent service record. Those with extraordinally low failure rates other than mechanical components (like disk drives). MOST of the problems with clones that have been shipped are caused by shipper abuse. We see it all the time - UPS is hell on computer systems, even when they are correctly packed. And it's mechanical motion parts, like disk drives, that take the brunt of it. You simply cannot blame that on the clone maker -- it is beyond his control (except after the fact, and you have a right to expect prompt attention to these problems!) Hell, Fed-Ex has trashed a machine or two on us before! MOST reasonable clone shops test their systems before shipment. Good places do at least a 72 hour burn-in, with some kind of automated software and/or hardware to ensure that everything is working right BEFORE it leaves. That is one of the reasons you will >never< get a system from our company that has "original" wrapping on the component parts (or get parts in a box). We check >every< piece, and to do that they all have to be installed and exercised -- as a unit. I'm not going to claim that all clone shops are this good. I know some aren't -- some are literally shlock operations, and some of the big guys don't give a tinker's damn about quality control either! In general the magazine "cheapie" places can't afford that kind of quality control, so they don't do it. Simple enough. You, the buyer, do it for them. Quality techs aren't cheap to employ! >The larger clone makers have their own IC design teams to design customer >circuits for their clones. The resulting clone may be faster, cheaper or more >reliable than a clone made entirely of off-the-shelf components. >Check the motherboards on the clone. The small clone makers buy their >motherboards off the shelf from an oem supplier. The large >clone makers make their own motherboards. A custom motherboard >is a good sign that this is not just another clone. A custom motherboard, power supply, or other components (such as HP's integrated "everything buy the kitchen sink boards" that control disk I/O, serial and parallel ports) also raise the cost of repair by an exhorbitant amount. What does a replacement motherboard cost for one of those ES12s? Or a QS16? Or, God Forbid, the plug-in module for the RS series systems? Now what does a "clone" motherboard cost? What do you pay to replace that blown serial port? $60 for a clone board? Or $300 for the "integrated schmaltz controller with everything" card? Once you're out of warranty these things become rather, uh, important. Cost of ownership includes much more than the original selling price. >I work for a clone maker testing new clones. Most of >what goes in our Vectras are custom designed by us. The stuff >we oem off the shelf has to meet our internal Strife standards. >InfoWorld wrote an article recently on the tests an oem component >has to pass to qualify before we will use it. You can bet a clone >maker operating out of a garage does not do this sort of testing. >You may be getting an equally good product (after all that is >how Hewlett-Packard got started) but that is unlikely. Is it unlikely? I'd be happy to dispute that. We service nearly every brand of PC in existance today. One of our customers has a lot of HP equipment due to political reasons (by their own admission). We love this -- their repair business is darn good to us, in no small part due to the number of Vectras on their site. You use Seagate Drives in the Vectras (ES12/QS16 models). I have replaced over 10 of them in the last year with "stiction" problems. Looks like HP quality control didn't catch it any better than anyone else's quality control! What did "Strife" do for you there? All of them have your own HP control sticker on them, and every one of them is junk. Just like a lot of other Seagates. Thanks to the warranty on some of those units (90 days!) the drives were out of warranty, and the customer got to eat the cost of the new disk drive. Most clone shops? One year warranty, parts and labor. Most notably, none of the replacements we have installed have failed. I've also replaced three of your "super-duper controller-I/O boards" in Vectra PC's. All out of warranty by a couple of months. All replaced with standard controllers and I/O cards, since that was a >lot< cheaper than replacing with "original" HP parts.... and again, all of those replacements are still working. Look at the parts in a Vectra. Paradise video card, for all intents and purposes (or at least Paradise video controller chips). Custom IC's? Where? On the motherboard, maybe. On the peripheral boards? Where? Custom >boards<? Yes. All over. Nice and expensive to replace, proprietary that they are. A good way to ensure (or try to) that you get the service business on the PCs, and get to charge whatever you want on the repair parts. At least HP doesn't etch the manufacturer's markings off the chips, like some other companies. This contributes to my being able to identify the disk controller as WD-based with some extra pieces tacked on for serial and parallel ports, the VGA video as a Paradise chipset, etc. Shall we talk about power supplies? ~105 Watts in the QS16's available to the user, while all the clone systems have 150 or 200 watt power supplies in that class? And before anyone says it's enough, it isn't -- we had power problems with a QS16. Don't put a full-height hard disk in a QS16 if you have additional memory and some cards in the system; there isn't enough juice there. 105 Watts is marginal for an XT, say much less any 80386 system! (Don't believe me? Go add up the nameplate ratings sometime and you'll see for yourself). Now, how much is a replacement power supply for a Vectra? You should have seen the red-faced person at Spring Comdex last year when I asked about the power supply rating in a QS16, he said "it's 150 watts", and I got him to take off the cover. It's not hard to multiply volts X amps and add the results....... 150 Watts drawn from the wall, perhaps. Delivered to the innards of the machine? No way! For all this "quality control" and "superior engineering" HP charges a premium price. They are not alone; Tandy and others do the same thing. They have a name, true. Does it ensure quality? Does the "quality control" do it either? You be the judge. Some firms test completed systems. I know we do, and I know factually that a few other firms do. I also know that lots of companies DO NOT; they ship machines in parts for "final assembly" by the user. How do they know whether or not the parts all work together, this particular set? The answer, of course, is that they don't. Clones are a good buy, as long as you get a system from a reliable source. Just remember, you cannot measure the cost of ownership (which is what really matters!) by either the name on the front or the price paid alone. You must look much deeper than that. The analysis is not easy, but if you care about the quality of what you purchase it must be done. Those who buy only on the basis of price are never going to get that kind of care in assembly and component choice. If you are willing to spend more, you can either spend a little more and find a clone shop that cares about these items, yet doesn't sacrifice the standard-component interchangability, or go with something like the Vectra (or Tandy) lines and spend lots of cash. The choice, as always, lies with the consumer. And the "top-of-the-line" name brands aren't always better, or even as good as the no-name clones. In many cases they are just different -- different enough to cost you should repairs be needed later on. An informed consumer is our best customer. Disclaimer: We service and sell PC Clones. -- Karl Denninger (karl@ddsw1.MCS.COM, <well-connected>!ddsw1!karl) Public Access Data Line: [+1 708 566-8911], Voice: [+1 708 566-8910] Macro Computer Solutions, Inc. "Quality Solutions at a Fair Price"
werner@aecom.yu.edu (Craig Werner) (03/11/90)
In article <1592@starfish.Convergent.COM>, tims@starfish.Convergent.COM (Tim Simmons) writes: > lane@cs.dal.ca (John Wright/Dr. Pat Lane) writes: > > >I wonder why computer mags and people here on the net spend so > >much time comparing various clone computer brands, most of them from > >from mail-order houses in the U.S. Surely all these brands are assembled > >from the same generic components that you can buy from these places. If that assumption were true, then why would the results of the reviews come out so different? They do. For instance, Dell comes out way on top of Fountain. Zeos gets high reviews. Most of the places are not just slapping together components. They are more like buying parts in volume and sending out complete systems. Very few of the Gateway/Dell/ Zeos/CompuAdd/Swan/etc. companies sell their parts separately any more. Most have grown up into the whole systems approach, and for that reason, reviews are meaningful! -- Craig Werner (future MD/PhD, 4.5 years down, 2.5 to go) werner@aecom.YU.EDU -- Albert Einstein College of Medicine (1935-14E Eastchester Rd., Bronx NY 10461, 212-931-2517) "Results would only confuse people."
francis@cunixf.cc.columbia.edu (Francis Ho) (03/12/90)
In article <1990Mar9.062402.22830@agate.berkeley.edu> ilan343@violet.berkeley.edu writes: > >I agree with the main thread of the original posting. Most of the >parts in these clones are generic components. Other than price and >support, there is very little difference between "Joe's down the >corner" and the higher-volume mail-order houses. there is still such a thing called the FCC rating and believe it or not, it does matter. /f $**********$****************$*************************************$**********$ * Francis Ivanos Ho USnail | 531 West 113th St. NYC 10025 * \ / * *____/\____* bitnet | francis@cunixc.bitnet * [__] * * / \ * internet | francis@cunixF.cc.columbia.edu * v \oo/ * * /<()>\ * UUCP | ...!rutgers!columbia!cunixc!francis * `/#/-c * * /______\ * Columbia Center for Computing Activities (CUCCA)... * _/ \_ * $**********$****************$************************************************$ --I just want to be me so I speak for myself.
mvolo@uncecs.edu (Michael R. Volow) (03/13/90)
Not only are the larger clone guys (Swan, Zeos, Dell, etc.) selling whole systems. Even if you want to move your fast hard disk and VGA video out of your XT into a basic 286/386, you can't even buy th basic system from the bigger clone dealers anymore -- they insist on selling you a whole system. I guess they don't want the hassle of trying to help customers integrate dozens of different types of hard disks and videos with their systems. M Volow, VA Medical Center, Durham, NC 27705 mvolo@ecsvax.edu 919 286 0411
dlow@hpspcoi.HP.COM (Danny Low) (03/14/90)
>MOST of the problems with clones that have been shipped are caused by >shipper abuse. ... You simply cannot blame that on the >clone maker -- it is beyond his control This is NOT beyond the control of the clone maker. Proper packing can reduce this problem. Strife testing includes shaking and drop of the Vectra PC in shipping containers to find any problems in this area. >You use Seagate Drives in the Vectras (ES12/QS16 models). I have replaced >over 10 of them in the last year with "stiction" problems. Looks like HP >quality control didn't catch it any better than anyone else's quality control! We are not too happy about this problem either. If you are replacing stiction problem drives you are doing your customers a disservice. The problem can be solved by taking the drive out and giving it a quick twist counterclockwise to spin the disk free. This problem is inherent in the design of the drive according to a Seagate engineer I talked to. Replacing the drive is only a solution if you replace it with a non-Seagate drive. Danny Low HP SPCD dlow%hpspcoi@hplabs.hp.com ...!hplabs!hpspcoi!dlow
rzahavi@gateway.mitre.org (Ron Zahavi) (03/14/90)
In article <6829@hydra.gatech.EDU> gt0159a@prism.gatech.EDU (LEVINSON,MARC LOUIS) writes: > >I find that PC Magazine's reviews give a certain credibility to the >vendor/clonemaker's claims, or disprove them. No I don't take their word >for the absolute truth, but I do use it as one of the measures of >comparison. PLEASE NOTE: PC Magazine makes every effort to list the >exact components used in the box, therefore, tey review the parts as well. > I agree with what most have said about PC Magazine's reviews. Unfortunately PC Magazine often provides credibility when none is due, and vice versa. Case in point, a few years ago there was a clone maker in Northern Virginia that lucked out and sent in a good machine (probably spent hours making sure the thing worked properly). Because of their price and relative good construction, they were voted Editor's choice. The result was that the company was swamped with orders, their quality dropped and so did the company. I know of someone who purchased a machine one day, found out it didn't work and came back the next day to find that the company went bankrupt. My suggestion is - build your own PC, learn the components and provide your own warranty, or buy from someone that provides extra or 3rd party warranty. I have seen enough of the above cases to know it's worth the money. A buyer must ask him/herself the question "if the company is gone tomorrow what happens to my several thousand dollar investment? Was it worth the money I saved by buying a cheap clone?" -- Ron -- ============================================================================== Ron Zahavi (703) 883-5637 Mitre Corporation rzahavi@gateway.mitre.org 7525 Colshire Drive McLean, VA 22102 ==============================================================================
dlow@hpspcoi.HP.COM (Danny Low) (03/14/90)
>Look at the parts in a Vectra. Paradise video card, for all intents and >purposes (or at least Paradise video controller chips). Custom IC's? >Where? On the motherboard, maybe. On the peripheral boards? Where? Yes we use the Paradise VGA chip set. So do many other VGA board makers. However our VGA board is our own design and made in our own surface mount facility. If you read the various VGA comparison reports, the HP VGA cards passes/fails the compatibility tests differently than the VGA cards from Paradise. In that respect we are no different the other VGA makers who also use the Paradise chip set but design and build their own VGA cards. They also pass/fail the tests differently. Danny Low "Question Authority and the Authorities will question You" Valley of Hearts Delight, Silicon Valley HP SPCD dlow%hpspcoi@hplabs.hp.com ...!hplabs!hpspcoi!dlow
karl@ddsw1.MCS.COM (Karl Denninger) (03/15/90)
>>I wrote.... >Response 10 of 11 (9561) by dlow at hpspcoi.HP.COM on Wed 14 Mar 90 03:32 >[Danny Low] >(22 lines) > >>MOST of the problems with clones that have been shipped are caused by >>shipper abuse. ... You simply cannot blame that on the >>clone maker -- it is beyond his control > >This is NOT beyond the control of the clone maker. Proper packing >can reduce this problem. Strife testing includes shaking and drop of the >Vectra PC in shipping containers to find any problems in this area. I don't think you have understood my point -- or perhaps you don't do incoming inspection of cartons or have to deal with it when the Gorillas get ahold of your product. I'd like to see your Vectra, or ANYONES equipment survive some of what I've seen UPS do. When the system is reduced to 1/2 it's initial height, or someone forks the side of the box (with a forklift!), or the carton is crushed due to a 5+ foot fall onto hard concrete, then you are screwed, pure and simple. Shock limits for fixed disks are typically 25-40Gs instantaneous. Proper packing helps, but it will not solve the problem when the shipper drops the system from a height of 5+ feet, crushes it beyond recognition, or impales it with a forklift or other "interesting" objects! I have noted that you aren't one of the few makers who spend the extra money to put "shockwatch" devices on your PCs or shipping cartons....those positively identify devices which have been subjected to unreasonable shock loads. (In all fairness, I've never seen a clone makers do this -- but DEC does on many of their drive devices, including the old "RL" removable disk packs!) >>You use Seagate Drives in the Vectras (ES12/QS16 models). I have replaced >>over 10 of them in the last year with "stiction" problems. Looks like HP >>quality control didn't catch it any better than anyone else's quality >>control! > >We are not too happy about this problem either. If you are replacing >stiction problem drives you are doing your customers a disservice. How? The system was down, and now it's working. And it has stayed working. >The problem can be solved by taking the drive out and giving it a >quick twist counterclockwise to spin the disk free. This problem >is inherent in the design of the drive according to a Seagate engineer >I talked to. Replacing the drive is only a solution if you replace it >with a non-Seagate drive. I hate to say it, but you don't know what you are talking about. We've talked to companies without a vested interest -- several repair firms that fix these things -- and we have done our own inspection of failed/dead drives. You >can< free up the drive by doing the "twisties", or by <gently> moving the stepper motor shaft. But that doesn't solve the problem, it just makes it work this time. A week, or a day, or an hour later, when the system is again powered down and left to cool, it may stick again. Given enough time, it WILL stick again. Once these drives show this symptom, the only permanent fix is replacement. The problem is not inherent to the design of the drive, it is (as has been discussed here before) a problem with the lubricant on the platters. Specifically, some drives had too much lubricant applied in the manufacturing process, which in operation ends up in the "park" area. This stuff is viscous enough that coupled with the head design, and a sufficient quantity of it in one place, the heads are prevented from moving away from "park" when power is applied. This results in a "drive unsafe" status and the spindle refuses to start rotating. The only true fix is replacement of the drive or a rebuild -- and a rebuilt drive doesn't have a new warranty. We give the customers the choice, and so far all of them have opted for replacement. Note that it has nothing to do with the spindle motor itself. If it did, then rotating the stepper shaft would not "unstick" the drives - but it does. We have also seen spindle motors that have a "blown" winding -- these exhibit many of the same symptoms. Again, the only "fix" is replacement. If it was inherent in the design we'd still be getting stiction problems, and we're not. I haven't had a "stuck" drive now for over six months in systems that we have shipped. Seagate seems to have solved the manufacturing QC problem......perhaps they changed the lubricant formulation, or did something else. Who knows. All we do know is that the problem seems to be gone. My point was simply that your firm's quality assurance/quality control programs didn't do any better than anyone else in this specific case. Is that an indictment of HP? Nope. It does, however, say that your "Strife" program failed to catch a major QC problem with a vendor. Everyone else got nailed too, we replaced a half-dozen of these drives under warranty (thank the Gods that all the trouble showed up before the 1 year warranty expired on our machines!) >>Look at the parts in a Vectra. Paradise video card, for all intents and >>purposes (or at least Paradise video controller chips). Custom IC's? >>Where? On the motherboard, maybe. On the peripheral boards? Where? > >Yes we use the Paradise VGA chip set. So do many other VGA board >makers. However our VGA board is our own design and made in our >own surface mount facility. If you read the various VGA comparison reports, >the HP VGA cards passes/fails the compatibility tests differently >than the VGA cards from Paradise. In that respect we are no >different the other VGA makers who also use the Paradise chip set >but design and build their own VGA cards. >They also pass/fail the tests differently. Correct. The point is that it is awfully presumptious to say "We're better" when you use the same parts as everyone else! You may make the board yourself, but let me ask you -- when was the last time you had a >board< fail? The board itself, not the components.... And what about the cost of repair to the customer? I noticed that wasn't addressed. For most customers it is very important, especially once the warranty expires. This is the place where most of the "name brands" fall down; the proprietary components are a real burden once you are out of the factory-warranty protection. -- Karl Denninger (karl@ddsw1.MCS.COM, <well-connected>!ddsw1!karl) Public Access Data Line: [+1 708 566-8911], Voice: [+1 708 566-8910] Macro Computer Solutions, Inc. "Quality Solutions at a Fair Price"
dlow@hpspcoi.HP.COM (Danny Low) (03/16/90)
>It does, however, say that your "Strife" program failed to catch a major QC >problem with a vendor. Everyone else got nailed too, we replaced a >half-dozen of these drives under warranty (thank the Gods that all the >trouble showed up before the 1 year warranty expired on our machines!) Strife does not catch quality problems when a supplier screws up and sends us a bad batch of parts. That is not the purpose of Strife. Strife tests whether the NORMAL quality of the parts meets our reliability standard. Determining whether the evaluation units sent to us is normal is an interesting problem in itself. Qualifying a supplier is sometimes like buying a used car. >Correct. The point is that it is awfully presumptious to say "We're better" >when you use the same parts as everyone else! You may make the board >yourself, but let me ask you -- when was the last time you had a >board< >fail? The board itself, not the components.... Our boards are very reliable according to our warranty data. Also HOW you design and manufacture the boards makes a great of difference in the failure rate of the components in the boards. Place a screw hole too close to an IC trace and you get a dramatic increase in field failures. Analyzing why a component fails is very interesting. More often than not, it is not a problem with the component but usually with the design or manufacturing process of the larger part. Solder shorts, cross talk between traces that were too close together, excessive flexing of the board because a large part was placed in the middle of the board, etc. >And what about the cost of repair to the customer? I noticed that wasn't >addressed. For most customers it is very important, especially once the >warranty expires. Haven't you heard? HP really stands for High Priced. We do not design for lowest cost. We design for lowest cost that meets our quality standards. Sometimes (as in the Laserjet 2P) we just happen to be the lowest cost product on the market. Usually we are not. Being high priced certainly has not hurt us. The public records on our growth as a corporation and as a PC supplier show we are a very fast growing company. There are people who are willing to pay the price for quality. There are others for whom low price is more important. Whatever your perception of HP quality and price, there are many people out there whose perception is that HP's quality is worth the price. We do not claim perfection. We have certainly put out our share of lemons. However our practices are designed to minimize this. That's the whole point of my original posting. Different clone makers have different practices. This makes a difference in the likelihood of getting a reliable unit. Everyone can agree that more reliable is better. Cost is more subjective. What is cheap to one person is too expensive for another. Danny Low HP SPCD dlow%hpspcoi@hplabs.hp.com ...!hplabs!hpspcoi!dlow
boyne@hplvli.HP.COM (Art Boyne) (03/17/90)
karl@ddsw1.MCS.COM (Karl Denninger) writes: >>Response 10 of 11 (9561) by dlow at hpspcoi.HP.COM on Wed 14 Mar 90 03:32 >>[Danny Low] >> >>We are not too happy about this problem either. If you are replacing >>stiction problem drives you are doing your customers a disservice. >>This problem >>is inherent in the design of the drive according to a Seagate engineer >>I talked to. >Given enough time, >it WILL stick again. Once these drives show this symptom, the only permanent >fix is replacement. Sorry, Danny, but Karl is correct according to information I received from engineers at HP's Greeley Division, which used to deal with Seagate drives. However, Karl's description of the problem, while basically correct, is not perfect. The problem in the past (eg., ST4051) has been with the seals between the spindle motor and the sealed media compartment. The lubricant *outgasses* through the seals, leaving a sticky residue on the inner tracks, which are indeed the "park" area. This is a long-term cummulative problem, and when enough residue has collected, the heads begin to stick and, while the "twist" will free them early on, eventually enough residue accumulates that you will tear the heads off the actuator trying to free them. Replacement *is* the only long-term solution. The Greeley Division engineers also told me that, while some drives will fail and others not, those destined to fail start showing symptoms *very quickly*. A drive that has no symptom after ~6 months are unlikely to ever fail. (Witness the ST4051 that I have, manufactured during the height of their problems, now 5 years old and going strong.) Art Boyne, boyne@hplvla.hp.com
karl@ddsw1.MCS.COM (Karl Denninger) (03/20/90)
>----- >Response 14 of 14 (9561) by boyne at hplvli.HP.COM on Sun 18 Mar 90 15:26 >[Art Boyne] >(34 lines) (lots on stiction problems with Seagate drives) Note too that I've never had a voice-coil Seagate drive fail in this manner, including 4051's, 4096's, etc. Only the 251's seem to be succeptible to this mode of failure (this includes the St277s, as they are the same drive but with RLL rating). -- Karl Denninger (karl@ddsw1.MCS.COM, <well-connected>!ddsw1!karl) Public Access Data Line: [+1 708 566-8911], Voice: [+1 708 566-8910] Macro Computer Solutions, Inc. "Quality Solutions at a Fair Price"
Ralf.Brown@B.GP.CS.CMU.EDU (03/20/90)
In article <260564e0-2559.15comp.ibmpc-1@ddsw1.MCS.COM>, karl@ddsw1.MCS.COM (Karl Denninger) wrote: }>Response 14 of 14 (9561) by boyne at hplvli.HP.COM on Sun 18 Mar 90 15:26 }>[Art Boyne] }(lots on stiction problems with Seagate drives) } }Note too that I've never had a voice-coil Seagate drive fail in this manner, }including 4051's, 4096's, etc. Only the 251's seem to be succeptible to Then you were lucky. I had a 4051 fail due to stiction after about 2.5 years (though the first incident was after about 17 months, when it had been turned off for two weeks while I was on vacation). -- UUCP: {ucbvax,harvard}!cs.cmu.edu!ralf -=- 412-268-3053 (school) -=- FAX: ask ARPA: ralf@cs.cmu.edu BIT: ralf%cs.cmu.edu@CMUCCVMA FIDO: Ralf Brown 1:129/46 "How to Prove It" by Dana Angluin Disclaimer? I claimed something? 16. proof by cosmology: The negation of the proposition is unimaginable or meaningless. Popular for proofs of the existence of God.
boyne@hplvli.HP.COM (Art Boyne) (03/22/90)
karl@ddsw1.MCS.COM (Karl Denninger) writes >>Response 14 of 14 (9561) by boyne at hplvli.HP.COM on Sun 18 Mar 90 15:26 >> >>(lots on stiction problems with Seagate drives) > >Note too that I've never had a voice-coil Seagate drive fail in this manner, >including 4051's, 4096's, etc. Only the 251's seem to be succeptible to >this mode of failure The problems with the ST4051 occurred in early 1985, if my memory serves me correctly. They were corrected after a few months of production. This drive was used in one of the HP 9133 series disk products (I believe it was the 9133H), and the stiction problem caused a lot of grief for the production engineers who were the source of my information. Art Boyne, boyne@hplvla.hp.com