High Quality VHDCI Cables

[Will]

I'm looking for a supplier of VHDCI cables that are high quality and
reasonable cost. For a 3 ft cable, I would define reasonable cost as $25
to $30/cable.

We are starting to use a Paralan SCSI Cable Tester religiously on all cables
we buy, and we are finding failures at LVD speeds on at least 60% of the
cables we are buying as new. Many of these cable resellers don't appear to
even have a SCSI cable tester in house. I would appreciate a reference to
a vendor with consistent high quality in my price target range.

 

[David Lesher]

I'm looking for a supplier of VHDCI cables that are high quality and
reasonable cost. For a 3 ft cable, I would define reasonable cost as $25
to $30/cable.

We are starting to use a Paralan SCSI Cable Tester religiously on all cables
we buy, and we are finding failures at LVD speeds on at least 60% of the
cables we are buying as new. Many of these cable resellers don't appear to
even have a SCSI cable tester in house. I would appreciate a reference to
a vendor with consistent high quality in my price target range.

I've bought some from Alliance Computer Exchange, 330-821-0500, but
I can tell you that they surely don't test them; they are an
excess/surplus house.

eBay store "assetbuyers" had a bunch of dell-excess cables as well.

 

[Will]

70% of the cables we have purchased on eBay, including brand-new-unopened HP
and Compaq cables, have failed testing by a SCSI tester. Since we test
tape libraries in some volume, we are very very sensitive to this issue, as
a single bad cable in a SCSI chain can result in bizarre unexplained errors
during a long backup test, and such errors can take a long time to diagnose.

At this point I am so fed up with the quality problems with eBay cable
product that I am looking for a vendor who knows how to test, and who
actively spot tests everything they sell.

--
Will

I'm looking for a supplier of VHDCI cables that are high quality and
reasonable cost. For a 3 ft cable, I would define reasonable cost as $25
to $30/cable.

We are starting to use a Paralan SCSI Cable Tester religiously on all cables
we buy, and we are finding failures at LVD speeds on at least 60% of the
cables we are buying as new. Many of these cable resellers don't appear to
even have a SCSI cable tester in house. I would appreciate a reference to
a vendor with consistent high quality in my price target range.

I've bought some from Alliance Computer Exchange, 330-821-0500, but
I can tell you that they surely don't test them; they are an
excess/surplus house.

eBay store "assetbuyers" had a bunch of dell-excess cables as well.



--
A host is a host from coast to (e-mail address removed)
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

 

[Rod Speed]

70% of the cables we have purchased on eBay,
including brand-new-unopened HP and Compaq
cables, have failed testing by a SCSI tester.

Then its likely the tester is ****ed.

Since we test tape libraries in some volume, we are very
very sensitive to this issue, as a single bad cable in a SCSI
chain can result in bizarre unexplained errors during a long
backup test, and such errors can take a long time to diagnose.

At this point I am so fed up with the quality problems with eBay
cable product that I am looking for a vendor who knows how to
test, and who actively spot tests everything they sell.

I dont believe that HP and Compaq dont.

I'm looking for a supplier of VHDCI cables that are high quality and
reasonable cost. For a 3 ft cable, I would define reasonable
cost as $25 to $30/cable.

We are starting to use a Paralan SCSI Cable Tester religiously on
all cables we buy, and we are finding failures at LVD speeds on at
least 60% of the cables we are buying as new. Many of these cable
resellers don't appear to even have a SCSI cable tester in house.
I would appreciate a reference to a vendor with consistent high
quality in my price target range.

I've bought some from Alliance Computer Exchange, 330-821-0500, but
I can tell you that they surely don't test them; they are an
excess/surplus house.

eBay store "assetbuyers" had a bunch of dell-excess cables as well.



--
A host is a host from coast to (e-mail address removed)
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

 

[Will]

Then its likely the tester is ****ed.

That's what we thought too, but we had Paralan physically disassemble each
of those cables, and they confirmed the results of the tester by visual
inspection and testing in their lab.

I dont believe that HP and Compaq dont.

Half of what ends up on eBay is what HP and Compaq rejects and that they
sell off through Dovebid or other auction houses as "surplus". You can't
assume that because anything is new in original packaging that it came
through an ordinary retail channel that controlled quality.


I've had good luck with the 4M+ lengths on the Dell cables but I have not
seen a 1M VHDCI to VHDCI Dell cable anywhere.

I would however point out that most of what comes on eBay for refurb Dell is
coming from Dell finance as "as-is" merchandise. A supplier of ours buys
millions from them each year and is always complaining to me about how items
they receive are smashed in, broken, etc, and that Dell never discloses that
well to potential buyers. So, again, I would rather not become the beta
tester for sellers, but would like to identify a supplier who bothered to
buy a SCSI cable tester and use it before they tried to sell the cable.
I'm willing to pay extra for each cable to have them outsource that function
so I can know with high certainty that anything I buy from them works for
LVD and ultra-320 applications.

--
Will

--
A host is a host from coast to (e-mail address removed)
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

 

[Rod Speed]

That's what we thought too, but we had Paralan physically
disassemble each of those cables, and they confirmed the
results of the tester by visual inspection and testing in their lab.

Quite frankly I dont believe that with the HP and Compaq cables.

I'm happy to accept that with the generic cables.

So what percentage of the HP and Compaq cables were bad ?

Half of what ends up on eBay is what HP and Compaq rejects

I dont believe they let those get out with their brands on them.

and that they sell off through Dovebid or other auction houses as "surplus".

Dont believe that.

You can't assume that because anything is new in original packaging
that it came through an ordinary retail channel that controlled quality.

Yes you can with what gets out with their brand on them.

Of course you might well be seeing some counterfeit products
instead. The obvious way to avoid that is to buy from HP, not eBay.

I dont believe that dell sells their rejects that way.

I've had good luck with the 4M+ lengths on the Dell cables but
I have not seen a 1M VHDCI to VHDCI Dell cable anywhere.

Sure, some of the counterfeits do work fine. Cables arent rocket
science and most branded stuff does come with chinese cables now.

I would however point out that most of what comes on eBay for
refurb Dell is coming from Dell finance as "as-is" merchandise.

Sure, but that doesnt mean that its tested rejects.

A supplier of ours buys millions from them each year and is always
complaining to me about how items they receive are smashed in,
broken, etc, and that Dell never discloses that well to potential buyers.

Sure, returns from customers are always higher risk.

So, again, I would rather not become the beta tester for sellers,
but would like to identify a supplier who bothered to buy a SCSI
cable tester and use it before they tried to sell the cable.

Thats what gets sold as refurbs, not what gets sold as returns.

I'm willing to pay extra for each cable to have them outsource
that function so I can know with high certainty that anything I
buy from them works for LVD and ultra-320 applications.

Thats what refurbs are.

 

[Folkert Rienstra]

70% of the cables we have purchased on eBay, including brand-new-unopened
HP and Compaq cables, have failed testing by a SCSI tester. Since we test
tape libraries in some volume, we are very very sensitive to this issue, as
a single bad cable in a SCSI chain can result in bizarre unexplained errors
during a long backup test, and such errors can take a long time to diagnose.

If the manufacture of SCSI cables is so sensitive that only 30% is suitable for
use, what makes you think that they stay that way while *in* use (and abused).

At this point I am so fed up with the quality problems with eBay cable product

And what exactly would be different if you had bought them from Compaq/Dell
etc., direct?

that I am looking for a vendor who knows how to test, and who actively spot
tests everything they sell.

The original manufacturer should have done that already.
One would only do that if one would sell some cheap product at enhanced value.

 

[Will]

Quite frankly I dont believe that with the HP and Compaq cables.

I'm happy to accept that with the generic cables.

So what percentage of the HP and Compaq cables were bad ?

I don't think it is a brand specific comment. It's an eBay comment. The
fact is a high percentage of the LVD cables we have purchased from eBay have
tested as bad.

Regarding the HP cables, I would be happy to send you by snail mail one of a
batch of 20 "new in original Compaq box" cables that were about 90%
defective. You are free to physically disassemble it and do whatever kind
of testing on it you like. It's not interesting though to battle your
personal opinions about what cables work and which don't against Paralan's
quality engineers actual test results with a physically disassembled cable.

Of course you might well be seeing some counterfeit products
instead. The obvious way to avoid that is to buy from HP, not eBay.

That much I agree with you. There are tons of counterfeits on eBay and
they are hard to detect.

Thats what refurbs are.

The problem is anyone can say an item is "refurb" or "tested", but when you
ask most eBay sellers of cables what is their "testing procedure", what you
can back is nonsense.

In any case, the purpose of this thread was to find a seller who a)
understands what a SCSI cable tester is; b) uses it; c) practices active
quality management of the LVD cables that they sell. I am happy to pay
extra for products from a seller who meets the purchase criteria above.

 

[Folkert Rienstra]

I don't think it is a brand specific comment. It's an eBay comment. The
fact is a high percentage of the LVD cables we have purchased from eBay
have tested as bad.

Regarding the HP cables, I would be happy to send you by snail mail one of
a batch of 20 "new in original Compaq box"

Oh well, HP cables in a Compaq box, what did you expect. ;-)

cables that were about 90% defective.

So, is that exactly 'about' 90% (18 out of 20) or exactly 85% (17 out of 20)
or more like exactly 95% (19 out of 20)?

Btw, sounds more like a faulty batch (say a new worker managed to do
'm all wrong) that found it's way through quality control undetected.

You are free to physically disassemble it and do whatever kind of test-
ing on it you like. It's not interesting though to battle your personal
opinions about what cables work and which don't against Paralan's

quality engineers actual test results with a physically disassembled cable.

Uh, how does one test a physically disassembled cable.
Since by disassembling it, you will have ruined it already.

That much I agree with you. There are tons of counterfeits on eBay and
they are hard to detect.

And if your example is to be taken seriously there are rejected lots too
that should have been destroyed but apparently were sold off instead.

 

[David Lesher]

70% of the cables we have purchased on eBay, including brand-new-unopened HP
and Compaq cables, have failed testing by a SCSI tester. Since we test
tape libraries in some volume, we are very very sensitive to this issue, as
a single bad cable in a SCSI chain can result in bizarre unexplained errors
during a long backup test, and such errors can take a long time to diagnose.

So, plan accordingly. Buy & test enough inexpensive cables to get the
numbers of tested-good cables you require. Given that I paid ~$5 each
for the ones I got.....

 

[Will]

If these were very expensive items, or if we were buying high volumes of
them, I guess I might look at that. But we need about 10 cables a month,
and I value my time more than I value the $200 in savings.

--
Will

70% of the cables we have purchased on eBay, including brand-new-unopened HP
and Compaq cables, have failed testing by a SCSI tester. Since we test
tape libraries in some volume, we are very very sensitive to this issue, as
a single bad cable in a SCSI chain can result in bizarre unexplained errors
during a long backup test, and such errors can take a long time to

diagnose.

So, plan accordingly. Buy & test enough inexpensive cables to get the
numbers of tested-good cables you require. Given that I paid ~$5 each
for the ones I got.....

--
A host is a host from coast to (e-mail address removed)
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

 

[Rod Speed]

If these were very expensive items, or if we were buying high volumes
of them, I guess I might look at that. But we need about 10 cables
a month, and I value my time more than I value the $200 in savings.

So buy them from HP and not ebay and then
you have recourse if you get bad cables.

70% of the cables we have purchased on eBay, including
brand-new-unopened HP and Compaq cables, have failed testing by a
SCSI tester. Since we test tape libraries in some volume, we are
very very sensitive to this issue, as a single bad cable in a SCSI
chain can result in bizarre unexplained errors during a long backup
test, and such errors can take a long time to diagnose.

So, plan accordingly. Buy & test enough inexpensive cables to get the
numbers of tested-good cables you require. Given that I paid ~$5 each
for the ones I got.....

--
A host is a host from coast to (e-mail address removed)
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

 

[Will]

So far what I have seen is two extremes, and I'm searching for a middle
ground.

Option 1) You can buy the cables on eBay at $5 to $10/each, in which case
very large numbers of them fail testing, even the ones packaged as "new".

Option 2) You buy the cables from HP at $50 to $90/each, or from higher end
suppliers and pay $49/each. Presumably those work. I don't assume they
do because it is not where we have been buying them.

I'm looking for Option 3): You buy the cables from a third party who
discounts them aggressively (say $29/each), and whose internal quality
standards dictate that they intensively test the cables they sell.

 

[Rod Speed]

So far what I have seen is two extremes, and I'm searching for a
middle ground.

Option 1) You can buy the cables on eBay at $5 to $10/each, in which
case very large numbers of them fail testing, even the ones packaged
as "new".

Option 2) You buy the cables from HP at $50 to $90/each, or from
higher end suppliers and pay $49/each. Presumably those work. I
don't assume they do because it is not where we have been buying them.

I'm looking for Option 3): You buy the cables from a third party who
discounts them aggressively (say $29/each), and whose internal quality
standards dictate that they intensively test the cables they sell.

Seems rather unlikely that there will be too many that do that
so you are using quite a bit of your time looking for what is unusual.

Is it worth the trouble for a saving of say $20 per cable ?

 

[Rod Speed]

Seems rather unlikely that there will be too many that do that
so you are using quite a bit of your time looking for what is unusual.

Is it worth the trouble for a saving of say $20 per cable ?

Particularly given the cost of what you are plugging into that cable.

Sounds like getting lost in the basics too me.

 

[Folkert Rienstra]

So far what I have seen is two extremes, and I'm searching for a middle
ground.

Option 1) You can buy the cables on eBay at $5 to $10/each, in which case
very large numbers of them fail testing, even the ones packaged as "new".

Option 2) You buy the cables from HP at $50 to $90/each, or from higher end
suppliers and pay $49/each. Presumably those work. I don't assume they
do because it is not where we have been buying them.

I'm looking for Option 3): You buy the cables from a third party who
discounts them aggressively (say $29/each), and whose internal quality
standards dictate that they intensively test the cables they sell.

Rotflol. In other words you're looking for the impossible.

 

[Will]

Rotflol. In other words you're looking for the impossible.

It takes us about 45 seconds to test either 68 pin or VHDCI SCSI cables with
a good SCSI tester, testing at both ends of the cable. Even if it took two
minutes, if your cost in the cable is $10 delivered, selling it at $29 you
have plenty of margin to pay for a two minute test.

 

[Folkert Rienstra]

It takes us about 45 seconds to test either 68 pin or VHDCI SCSI
cables with a good SCSI tester, testing at both ends of the cable.

And that is "intensively" in your book?

Even if it took two minutes, if your cost in the cable is $10 delivered,
selling it at $29 you have plenty of margin to pay for a two minute test.

You want someone to sell you a "High Quality" (ie expensive) cable for far
less than the 'adviced' price (ie at a discount) involving more work (eating
even more out of that already reduced margin) and all that because you
can't be bothered to invest a measly 10 minutes per month to do it yourself.

Ok, good luck to you.

 

[Will]

And that is "intensively" in your book?

I would settle for any level of testing. Some testing is better than zero
testing.

test.

You want someone to sell you a "High Quality" (ie expensive) cable for far
less than the 'adviced' price (ie at a discount) involving more work (eating
even more out of that already reduced margin) and all that because you
can't be bothered to invest a measly 10 minutes per month to do it

yourself.

We do test the cables. Where did I say I didn't want to test them? The
point is we test them now and get lots of failures. I want to find a
vendor whose product we also test, but that product works, and I would
settle for even just 90% of the time.