#Lost in the Hardware???
9 messages in this thread
Hi,
Thanks for the input. I understand the diiferance a lot better now but
the company that I buy my system boards from advertizes 8 32MHz EISA bus slots
for the 50MHz that costs $400.00 more then the 66DX2MHz with only two VLB and 6
ISA bus slots. They agree that the 50 is a lot faster in all cases. Two of my
friends have these different boards and from what I can tell the 50 looks to be
twice as fast as the 66. Another friend has a Zeos 66DX2 with two VLB slots
and it''s not as fast as either of the others. And he has had to replace the
monitor and system board in the first month and the thing crashes if you look
at it wrong<G>
So now that I understand a lot more why does this EISA bus look so much
faster while everyone here says it's not??????? If I didn't like to use 3ds
so much I wouldn't listem to you guys on this one<G>.
Andy
Andy,
I can say with a good deal of certainty that the 66dx2 will give faster
rendering times in 3ds than the 50dx. As for other applications, I don't know
but speed for 3ds would be my main buying criteria.
Andy,
If they are charging you $400 more for an EISA 50Mhz than for a VLB 66, then
they are ripping you off. I've seen 50MHz boards for less than motheboards
with the 66 cpu, and I've seen motherboards with EISA+VLB for only $150-200
more than ones with VLB only. Also, I'm sure he means 32 BITS not MHZ when
referring to the EISA slots.
Also, what brand of motherboards are these? If I were buying a motherboard,
I'd get only one of the top brands: AMI, Micronics, Mylex, Haupagge, Orchid,
etc. and not one of the off brands like Opti. You may do ok buying a complete
system with an off brand, but usually you can return a complete system.
The ZEOS is known to do poorly on the PC Mag memory speed tests. I think
whichever test you are running that show the 50 to be faster are internal cache
thrashing tests that would make the ZEOS look particularly bad. Also, my AMI
DX2/66 almost never crashes, certainly not in 3DS and rarely otherwise. I've
see lots of independent benchmarks that show that the DX2/66 is faster than the
DX/50 in most cases. Also, does the MB you are looking at come with a P24T
socket? If it does, then you will be able to upgrade to Pentium whenever Intel
releases the upgrade chip. The Pentium has such awesome floating point speed
compared to the 486 that it will be must-have.
Bailey
Hi,
You ask what MB's I buy, JDR. I wanted to buy a AMI but they wouldn't
sell me just the MB! Why JDR, I can get any brand drives,video, controllers,
or what ever and it's always compatible unlike some DELL, ZEOS. Also they have
24hour ordering, the best tech suport I've even seen with any company unlike
ZEOS that may call you back in a week or so, or DELL send a old monitor when
the two day old one blew up. I sell computers on the side using JDR MB's,
I''ve sold close to 500 of them and all of them are still working. I got
hooked on them when a friend that works for IBM told me that most of the
employees buy JDR because the cost is less and the quility is higher. There
also on line here GO JDR. Not tring to sell you one I'm just sold on them
after dealing with many junks.
If you say that I need a 66DX2 VLB for 3ds then I'll take my copy to my
friends houses and test out rendering on both machines. This is hard for me to
belive but I will let you know. Testing these computers before was not done
using any kind of testing program like checkit. I just timed how long it took
to load a large program like windows, the 50 could load and return to the menu
before the 66 got loaded. I tried several other programs with same results.
Well your probably right but I'll have to see for myself<G>.
Andy
PMJY, but… first of all, if you are so certain of JDR's quility and have sold
over 500 of them, why do you ask us, mere mortals? Second, what defines the
speed of loading up Windows is hard disk and hard disk controller. Even a Cray
would take longer to load Windows using a el cheapo HD+HDC (if such a thing
could be done. Windows would probably bend the Cray so out of shape that the
poor Cray would refuse to ever boot again). That's just not the right benchmark
procedure for testing CPU speed. The rendering test you proposed is the right
approach. As to everything else being junk… ain't that kinda presumptuous if
not just plain haughty?
Hi,
I agree that my tests have not been good ones, but at the same time my
customers want to see the differance not just have a testing program tell them
the CPU's performace. The reason I asked the quesstion is I want the best for
me and I WAS/am confused and want to give my customers good info on there
invesments.
I might have implied that all other computer where junk but Cmon that
was mostly humor. But there are some junks out there. Here are my estamated
statistats: 1 AMI computer, no trouble at all : 7 Zeos computers, all had to be
seviced in the first 6 months, and very lousy sevice at that : 3 Dells, 1
started on fire but pretty good service : Near 500 of mine which use JDR
systems boards, cases/PS, keyboard, Teac Floppys, NEC Monitors, Seagate SCSI
HD's, Hayes Modems. This is what I sell as a basic system and the only trouble
I've had is JDR mis-labled the RAM banks on one type system board. I'v sold
some no-name brands and there not much to say about them except don't buy one!
Ok now here are some tests I ran last night. Checkit reports that the
JDR 50Mhz EISA ran 1/3 faster as the Zeos 66MHz VLB in all areas except video
and math where there was not much differance, but the 50 was faster. Then I
renderd 30 frames to NULL on the 50 it took almost 1 1/2 hours, then on the 66
it took a little over 2 hours. Both systems only have 8 megs of RAM and have
ST3283N SCSI HD. Now what's going on here, I really expected you guys to be
right<G>. Can anyone uplaod a report from checkit from a 66 VLB so I can
compair it to these. Maybe this Zeos needs repair<G>.
Andy
Or maybe a sprinkler system<g>! I just got an Intel Pentium board to play with.
I guess I know what I'll be doing during the week-end.
Gus,
yeah! Let us know how it benchmarks with 3ds compared to a 486, please!
Andy,
will only 8mb RAM on those systems you may have been paging and that would
probably cause the 50 to be faster.