What do you make of this? Article discusses 3Gbps hard drives.

  • Thread starter Thread starter dg
  • Start date Start date
D

dg

http://channels.lockergnome.com/har...first_hard_drive_with_30_gbs_serial_ata.phtml

I have always claimed that the whole Ultra ATA 66/100/133 is just a scam, as
the interface is not the bottleneck. I get lots of people arguing with me
about that, probably because they don't want to believe they were taken by
the scam. Anyway, a co-worker of mine was just telling me about an article
he read about drives that have a 3Gbps transfer rate. Is this just another
marketing scam? If so, this is just getting ridiculous!

I always compare Ultra ATA 133 to putting Y rated tires on a VW bug (Y rated
means they can go 186 MPH without coming apart). Just because the tires can
go that fast doesn't mean the car can!

Thanks,
--Dan
 
Previously dg said:
I have always claimed that the whole Ultra ATA 66/100/133 is just a scam, as
the interface is not the bottleneck. I get lots of people arguing with me
about that, probably because they don't want to believe they were taken by
the scam.

Depends. With two modern disks on one bus ATA 66 is the bottleneck.
Also ATA has the problem that switching from one disk to another on
the same bus is slow.
Anyway, a co-worker of mine was just telling me about an article
he read about drives that have a 3Gbps transfer rate. Is this just another
marketing scam? If so, this is just getting ridiculous!

Well, the interface itself is cheap and whther it is 150MB/s or
300MB.s does not really matter. And if it is fast you will not
have to re-design it every few years. Look at ATA133. A modern
IDE disk can deliver >50MB/s. That is already pretty close to
problematic. Busses work best it they are not operated with their
maximum speed.
I always compare Ultra ATA 133 to putting Y rated tires on a VW bug
(Y rated means they can go 186 MPH without coming apart). Just
because the tires can go that fast doesn't mean the car can!

Not fair. The tires have a safety margin. They will likely
not come apart at 200MPH either. However there is no way in
this universe to get more than 133MB/s over ATA133 and
the practical limit is more likely 80-100MB/s.

Arno
 
The biggest problem is that many people think they are going to be
reading & writing all their files at 66, 100, 133MB/sec. It's
misleading to naive users. So are disk capacities, CPU MHz, etc.
Depends. With two modern disks on one bus ATA 66 is the bottleneck.

Doesn't it also depend on usage patterns?
Also ATA has the problem that switching from one disk to another on
the same bus is slow.

But the UDMA spec is supposed to support multithreaded IO. SATA
doesn't have to deal with this (at least not per/channel).
Well, the interface itself is cheap and whther it is 150MB/s or
300MB.s does not really matter. And if it is fast you will not
have to re-design it every few years.

SATA only supports 1 drive/channel so the ceiling is significantly
higher than PATA.
Look at ATA133. A modern
IDE disk can deliver >50MB/s. That is already pretty close to
problematic. Busses work best it they are not operated with their
maximum speed.

Is it really? that 50+MB/sec figure applies to serial reads along the
outer platter rings. With all the latency limitations, etc they
certainly never push files any where near that rate between drives. I
don't believe metadata take up most of the bandwidth.
Not fair. The tires have a safety margin. They will likely
not come apart at 200MPH either. However there is no way in
this universe to get more than 133MB/s over ATA133 and
the practical limit is more likely 80-100MB/s.

Arno

The point is the tires are overkill for the car. It has nothing to do
with the accuracy of the speed rating of the tires or whether it truly
is a maximal value.

I agree, though, when you say "Busses work best it they are not
operated with their maximum speed."
 
I see an application for this. An external disk array connected through such
interface.

Another consideration: serial SCSI uses the same physical layer, and if it
gets 3 Gbps mode, why not use the same PHY for SATA, too?
 
Previously Alexander Grigoriev said:
I see an application for this. An external disk array connected
through such interface.

A good example.
Another consideration: serial SCSI uses the same physical layer, and if it
gets 3 Gbps mode, why not use the same PHY for SATA, too?

Are you sure? I thought serial SCSI was using different dignal levels
(much larger) and supported several meters in cable length? AFAIK
serial SCSI can switch back to SATA signal levels but will then be
limited to 1 meter also. Do I have this wrong?

Arno
 
Arno said:
A good example.


Are you sure? I thought serial SCSI was using different dignal levels
(much larger) and supported several meters in cable length? AFAIK
serial SCSI can switch back to SATA signal levels but will then be
limited to 1 meter also. Do I have this wrong?

It's not clear what exactly they're talking about. Marvell doesn't have a
datasheet up on their site and the only thing Samsung has is a press
release. What exactly does 3 Gb/sec "transaction processing" mean in the
context of an embedded disk controller?
 
Previously J. Clarke said:
Arno Wagner wrote:
It's not clear what exactly they're talking about. Marvell doesn't have a
datasheet up on their site and the only thing Samsung has is a press
release.

I see. This means no real info yet.
What exactly does 3 Gb/sec "transaction processing" mean in the
context of an embedded disk controller?

Maybe a fancy name for "raw maximum speed"?

Arno
 
Arno said:
I see. This means no real info yet.


Maybe a fancy name for "raw maximum speed"?

Could be. "Transaction processing" usually denotes one of the journalling
schemes that database engines use--it's possible that the drive has some
kind of hardware support for that sort of thing.
 
Previously J. Clarke said:
Arno Wagner wrote:
Could be. "Transaction processing" usually denotes one of the journalling
schemes that database engines use--it's possible that the drive has some
kind of hardware support for that sort of thing.

I doubt that. More likely that are "bus transactions" meaning
a single data block transfer.

Arno
 
Curious George said:
The biggest problem is that many people think they are going to be
reading & writing all their files at 66, 100, 133MB/sec.
It's misleading to naive users.

Naive users mislead themselfs.
Thats probably why one calls them 'naive' in the first place.
So are disk capacities, CPU MHz, etc.

Only in RAID or with drives with a STR of over 60MB/s.
Doesn't it also depend on usage patterns?

Access pattern.Yup.
But then for the time NOT spend in seeks, data is still transferred slower.
It depends on the proportion of seek vs transfer time how much impact that
will have on the average transfer rate.
But the UDMA spec is supposed to support multithreaded IO.

There is no UDMA spec.
There is an ATA spec that started to cover Overlapped IO and Queueing.
SATA doesn't have to deal with this (at least not per/channel).

In theory. In practice this depends on the Host controller.

Of course it does.
Technological progress has to be fought for and comes at a price.
The price has to suit the product or it won't be bought.

Yet that is what you will be forced to do when technological progress
only allows you just so much and not more.
SATA only supports 1 drive/channel so the ceiling is significantly
higher than PATA.

Actually, that depends on protocoll overhead. We may have to wait to see
in practice how much that really is when the faster drives with SATA300
arrive and see what STR remains when connected to a SATA150 controller.
ATA overhead is ~10%. On top comes the serial protocol.
That is already pretty close to problematic.

About 9MB/s data bandwidth ceiling remaining.

Nonsense. That's applies to networks, not point to point.
Is it really? that 50+MB/sec figure applies to serial reads along the
outer platter rings.

Yes and? You don't want them let go to waste, don't you?
With all the latency limitations, etc

What latency limitations.
STR includes latency (head switches, cylinder switches).
The single track speeds are even higher.
they certainly never push files any where near that rate between drives.

If big enough and contiguous, yes they do.
I don't believe metadata take up most of the bandwidth.
Huh?


The point is the tires are overkill for the car. It has nothing to do
with the accuracy of the speed rating of the tires or whether it truly
is a maximal value.

I agree, though, when you say "Busses work best it they are not
operated with their maximum speed."

Then you are as clueless as he is.
A bus operates at the same speed all the time. Data is transferred in bursts.
Not running an ATA133 bus at it's maximum speed is running it at ATA100.
 
dg said:
http://channels.lockergnome.com/har...first_hard_drive_with_30_gbs_serial_ata.phtml

I have always claimed that the whole Ultra ATA 66/100/133 is just a scam,

So either you are very ignorant or just a TROLL.
as the interface is not the bottleneck.

The interface is never supposed to be the bottleneck.
As soon as the maximum capacity of an interface is starting to get
reached a new addition to the interface is introduced with enough
headroom available to last a few years and then it starts all over
again until the technology hits a barrier that isn't easily overcome.
If that happens that technology is abandoned and a new technology
is introduced.
The new technology is serial and it is point to point. Buses are out.

ATA is a 2-device bus, it has to support 2 devices simultaniously,
using time slicing, meaning that the bus must be twice as fast as a
single device can deliver so that by using half the bandwidth of the
channel, both can share it without loosing any data transfer speed.
I get lots of people arguing with me about that,

And that doesn't tell you something.
probably because they don't want to believe they were taken by the scam.

Yeah, they obviously can't be right.
You would have to kill yourself to save yourself from total embarrassment.
Anyway, a co-worker of mine was just telling me about an article
he read about drives that have a 3Gbps transfer rate.
Is this just another marketing scam? If so, this is just getting ridiculous!

Logic does not appear to be your strong point, is it.
If it was a scam before, you have to continue that scam, otherwise you concede
that it *was* a scam.
I always compare Ultra ATA 133 to putting Y rated tires on a VW bug (Y rated
means they can go 186 MPH without coming apart). Just because the tires can
go that fast doesn't mean the car can!

Your analogy is obviously flawed.
Using your analogy, a single VW Bus is delivering a maximum payload to A AND B
in the same time needed to deliver a single payload to either A OR B the normal way,
by catapulting it away at twice the normal speed. There's your 'Y rated tyres'.
By doing it in half the time for A it can do the same in the other half for B and
it will appear like it did it for both at the same time.
 
Folkert Rienstra said:
scam,

So either you are very ignorant or just a TROLL.

Ok, tell me which hard drives can exceed 66MBps transfer rate? How about
100MBps? 133MBps? When a person walks into a store and sees hard drives
labeled as if they transfer data at 133MBps, do you think they really
transfer data that fast? I know they can't transfer data nearly that fast,
thats why I feel it is a scam. Marketing.
The interface is never supposed to be the bottleneck.

So we agree.

--Dan
 
Naive users mislead themselfs.
Thats probably why one calls them 'naive' in the first place.

Its no secret the industry intentionally misleads buyers. You'll
never see usefull/adequate performance information when you go to buy
a new PC or from a disk manufacturer when you look up disk specs.
Drive transfer rates, if available at all, tend to be buried in
manufacturer specs, and like every other performance figure, the truth
is stretched. It takes quite a bit of sophistication to know what
matters and what you need when it comes to performance attributes.
Only in RAID or with drives with a STR of over 60MB/s.


Access pattern.Yup.
But then for the time NOT spend in seeks, data is still transferred slower.
It depends on the proportion of seek vs transfer time how much impact that
will have on the average transfer rate.

mentioned speed loss is the case of "RAID or with drives with a STR of
over 60MB/s" on ATA-66 but not most scenarios and usage patterns of
"two modern disks on one bus ATA 66."

The reality is that ATA 66 is fairly antiquated at this point and new
ata controllers are very cheap. It's purely academic to talk about
minor performance loss, or performance loss in a specific situation,
when connecting the newest drive you can find to the oldest controller
you can find. It's no secret that it is generally safe to assume that
mixing very old with very new has the potential to be suboptimal.
There is no UDMA spec.
There is an ATA spec that started to cover Overlapped IO and Queueing.

These features date back to ATA-ATAPI 4 which was dubbed "Ultra
DMA/33." Subsequent PATA standards also have UDMA marketing names.

If memory serves, multithreaded IO was the term coined to described a
group of features whose net effect were what we are parsing words to
try to describe.
In theory. In practice this depends on the Host controller.

Right, hence the caveat.
Of course it does.
Technological progress has to be fought for and comes at a price.

Fighting for technological progress by being an early adopter is
fighting windmills.
The price has to suit the product or it won't be bought.

And it has to seem shiny and new and either have more bells & whistles
or seem like it could be faster than the next guy's product.
Yet that is what you will be forced to do when technological progress
only allows you just so much and not more.

But _when_ will this technological limitation come?

Pushing technology is also part of how the business of technology
(including marketing) is done. It is not a matter or pure
engineering.

Unless it is incredibly inefficient, or disk manufacturers make
unprecedented strides, present SATA and SATA2 is in no imminent danger
of being inadequate. Of course we all expect it will eventually.
Actually, that depends on protocoll overhead. We may have to wait to see
in practice how much that really is when the faster drives with SATA300
arrive and see what STR remains when connected to a SATA150 controller.
ATA overhead is ~10%. On top comes the serial protocol.

Most PATA drives and mobos worth buying are ATA/100. I'm not
accustomed to seeing overhead >20 or 25% for anything.
About 9MB/s data bandwidth ceiling remaining.


Nonsense. That's applies to networks, not point to point.


Yes and? You don't want them let go to waste, don't you?

They won't really in many scenarios. Point is, though, they also are
not the "norm" across the whole disk. Once you install the OS, for
example, they are rarely utilized.
What latency limitations.
STR includes latency (head switches, cylinder switches).
The single track speeds are even higher.

But data tends not to be written as nicely as this. Try transferring
files you haven't handpicked for experiment and look at the average
file transfer rate to another identical disk. That's what end-users
really want to know when they look at external transfer rates.
Playing with numbers and scenarios is misleading to real, practical
concerns & questions.
If big enough and contiguous, yes they do.

Ahh- you see. File size and usage patterns again muddy the waters.
exactly.


Then you are as clueless as he is.
A bus operates at the same speed all the time. Data is transferred in bursts.
Not running an ATA133 bus at it's maximum speed is running it at ATA100.

No you misunderstand & are making inaccurate assumptions. We are
still talking about the relationship between the bus speed and the
speed of the attached drives and at what point is the bus speed
overkill and at what point is it inadequate for the drives attached.

I was referring to a situation where the drive/bus bandwidth rates are
close and if you are counting on say ATA-66 to support 66MB/sec
without taking into account the bandwidth lost to overhead, there is a
problem.
 
J. Clarke said:
It's not clear what exactly they're talking about. Marvell doesn't have a
datasheet up on their site and the only thing Samsung has is a press
release.
What exactly does 3 Gb/sec "transaction processing" mean in the
context of an embedded disk controller?

Nothing when expressed in bit, the smallest unit of data that doesn't
even have a meaningful purpose until combined to 8 or 10 bits.

Probably a marketing idiot that decided that 3 Gb/s sounded more impressive than 300 MB/s and changed it.

Maybe 'transactions' comes from SCSI where some SCSI makers use
MT/s (MegaTransactions which is basically MegaHertz) in stead of MB/s.
MT is narrow/wide independent which MB/s is not.
20MT/s denotes Ultra SCSI and is 20MB/s in narrow and 40MB/s in Wide.

But since serial is 1-bit wide and although bits/s is basically the clock it
isn't really helpful to call them 'transactions'.

 
Curious George said:
Its no secret the industry intentionally misleads buyers. You'll
never see usefull/adequate performance information when you go to buy
a new PC or from a disk manufacturer when you look up disk specs.

Well, that isn't exactly "intentional mislead".
Performance information isn't the same as specs, and specs is what they supply.
Drive transfer rates, if available at all, tend to be buried in
manufacturer specs, and like every other performance figure, the
truth is stretched.

If you read the specification manuals for diskdrives you'll find that the
performance figures usually comply whith those from the storage sites.
It takes quite a bit of sophistication to know what
matters and what you need when it comes to performance attributes.

Right. It takes an un-naive look.
mentioned speed loss is the case of "RAID or with drives with a STR of
over 60MB/s" on ATA-66 but not most scenarios and usage patterns of
"two modern disks on one bus ATA 66."

Since an ATA bus has to support two devices simultaniously, yes,
an ATA66 bus will support a single ATA133 drive just fine.

I thought you were arguing that access pattern would destroy the
average STR and that a lower busspeed therefor would still suffice.
That is only partially true.
The reality is that ATA 66 is fairly antiquated at this point and new
ata controllers are very cheap. It's purely academic to talk about
minor performance loss, or performance loss in a specific situation,
when connecting the newest drive you can find to the oldest controller
you can find. It's no secret that it is generally safe to assume that
mixing very old with very new has the potential to be suboptimal.


These features date back to ATA-ATAPI 4 which was dubbed "Ultra
DMA/33." Subsequent PATA standards also have UDMA marketing names.

Not in this group.
If memory serves, multithreaded IO was the term coined to describe a
group of features whose net effect were what we are parsing words to
try to describe.


Right, hence the caveat.

I think not. If the controller is a standard ATA controller
with sata->pata bridges added there is still the same problem.
Fighting for technological progress by being an early adopter is
fighting windmills.


And it has to seem shiny and new and either have more bells & whistles
or seem like it could be faster than the next guy's product.


But _when_ will this technological limitation come?

Pushing technology is also part of how the business of technology
(including marketing) is done. It is not a matter or pure engineering.

Unless it is incredibly inefficient, or disk manufacturers make
unprecedented strides, present SATA and SATA2 is in no imminent danger
of being inadequate. Of course we all expect it will eventually.


Most PATA drives and mobos worth buying are ATA/100. I'm not
accustomed to seeing overhead >20 or 25% for anything.

Read my lips: "On top comes the serial protocol".
And yes, SCSI protocol has about 25% overhead.
They won't really in many scenarios. Point is, though, they also are
not the "norm" across the whole disk. Once you install the OS, for
example, they are rarely utilized.


But data tends not to be written as nicely as this. Try transferring
files you haven't handpicked for experiment and look at the average
file transfer rate to another identical disk. That's what end-users
really want to know when they look at external transfer rates.
Playing with numbers and scenarios is misleading to real, practical
concerns & questions.

Yet the resultant average transfer rates are still a function
of the sustained or single track transfer rates. Limiting them
also limits the average transfer rates although not linearly.
Ahh- you see. File size and usage patterns again muddy the waters.


No you misunderstand & are making inaccurate assumptions.

I just read what is written there, and it was false.
We are
still talking about the relationship between the bus speed and the
speed of the attached drives and at what point is the bus speed
overkill and at what point is it inadequate for the drives attached.

Running a 60MB/s drive on an ATA66 bus is just fine. This will leave
no empy burst space on the bus, hence 100% (data) speed usage.
I was referring to a situation where the drive/bus bandwidth rates are
close and if you are counting on say ATA-66 to support 66MB/sec
without taking into account the bandwidth lost to overhead, there is a
problem.

Right, which is quite different to
"Busses work best it they are not operated with their maximum speed."
 
Well, that isn't exactly "intentional mislead".
Performance information isn't the same as specs, and specs is what they supply.

Only what a buyer needs/wants to make a purchase decision is
"performance information" or "specs _&_ performance info." This is
because buyers look at advertised info trying to determine what their
experience with the machine will be like and how it compares to a
competing product. Therefore in absence of any real, readily
digestible performance info buyers tend to try to draw inferences
about performance based on the specs. Resellers & manufacturers are
counting on this or for the buyer to be confused entirely by the
difference. Why? Because the market is so competitive, and profit
margins are frequently so slim, that even steering or stretching the
public perception about the benefits of a product a little bit can
make a big difference. Also if your competitor is doing it, it is
suicide to not play the same game.

When you buy a car, for example, and you want to know how fast it
goes. There are specs like "400 hp @ 6000 Rpm, 6.0 L V-8" but also
performance information like "0-60mph in 4.2 sec and top speed of 186
mph." Furthermore, it is easy for a customer to validate
manufacturer's claim. Testing your new car's acceleration & fuel
efficiency is very easy. Doing this, or at least feeling that you
could do this, helps give a sense of security that they "made the
right decision." What exactly is a customer supposed to do with
"ATA-100, 7200rpm" when they take it home? The info may be accurate
but it leaves you feeling uneasy and insecure. "Maybe I should've
bought SATA or ATA-133?" There is both the perception and the reality
that there is a lot of room to fudge a products assets & limitations
when _only_ handpicked specs are cited.

PC buyers don't generally get both specs and performance information
or if they do they have to convert the number, divide by a compression
ratio, recognize it's an artificial/maximal figure, etc. People are
used to shopping for cars, and even the most clueless drivers
understand a decent amount of the car terminology. The big problem is
PC buyers' expectations are wrong/misguided as they are based on other
kinds of purchases & products they already know. Yes carmakers trump
up their numbers too, partly through artificial testing, and also try
to confuse the un-savvy, but the buying experience as a whole is very
different due to the issues already described. (and yes it is
intentional that I cite car sales to be much more reputable than
computer sales).
If you read the specification manuals for diskdrives you'll find that the
performance figures usually comply whith those from the storage sites.

Yes and no. Even in a best case/most honest scenario specs are
exaggerated. First of all cited above are generally artificial,
maximal values. Also usually Mb/sec are used to create a bigger
number hoping that naive readers will think it is the same as MB/sec.
The numbers also do not directly translate to how files are going to
moving (average rates) once the drive is formatted and say OS
installed on it. Often you are not comparing apples to apples esp.
when all they see is a short list of specs like "80gig, ATA-100,
7200RPM." Other disk performance specs are even worse like drive
acoustics. Because of differing measuring methodologies you simply
can't compare the acoustic spec of say a Maxtor to a WD or Seagate.
And if you are in the market for a new PC, it's usually a total
crapshoot how noisy it will be.

I understand that there just isn't any widely accepted,
straightforward, unimpeachable performance measurement for them to
use, but the idea that the way tech items are described isn't
deliberate and part of shrewd marketing strategy is absurd. I'd argue
that to claim to the contrary is itself _extremely_ naive. I know,
Folkert, you have this notion that anyone that doesn't have your
perspective and knowledge is an idiot, but there is a larger world
than your own. The point of this thread is to discuss the issues
raised by the original poster. Your disdain for the average Joe is
irrelevant.
Right. It takes an un-naive look.

Exactly my point.

I'd continue to answer your objections point by point but I simply
don't have the time. I also don't think it's worthwhile as we aren't
generally disagreeing on _technical_ issues. Instead, all I'd be
doing is re-explaining my responses and voicing objections that you
are again loosing sight of the core argument of the thread or making
false assumptions about previous posts. I know how obsessed you are
with getting the last word and making the appearance of discrediting
someone even if your answers don't directly do so and/or stray from
the core arguments or even the point you are trying to contradict. So
go ahead. Say something snotty, have your last word, and lets put
this one to bed or at least hear someone else chime in.
 
One final thought re using raw specs like bus speed to describe a
product: it plays perfectly to the time honored marketing ploy/GM
mantra "Keep the Consumer Dissatisfied." I believe it was Kettering
who coined this phrase to describe his new marketing strategy that
included at its center holding back on features, R&D, and most
importantly model years for cars. The object was to entice consumers
to continue to buy newer model cars because of their actual or implied
improvements rather than waiting until consumers' current cars failed
to meet their needs. This is a core-marketing philosophy of virtually
all viable product-based industries.

Hard drives are generally sold to consumers using the specs: size,
spindle rpm, bus speed, and more recently cache size. None of these
specs really do justice to describing the complex nature of
"performance", nor do they reveal the tremendous performance range of
disks that share some or all of these specs. It instead creates a lot
of room to augment product perception and serves to create a product
scale akin to model years. ("The 2005 mustang must be improved over
the 2004 model"- just as "an ata-100 drive must be faster than an
ATA-66 one" or a "10K raptor must be about a third faster than a
7200rpm drive") It also creates potential upgrade targets in
consumers' minds, both when they buy and when they hear of new
technologies. It makes it easy to create a baseless illusion of
gains.

For the most part ATA disks exist almost exclusively as 1 or 2 per
computer and the computer as a whole primarily idles (that applies to
the enthusiast market as well). We have already discussed how in a
single drive/channel or one drive/channel used at a time ATA-66 is
absolutely fine. Yet ATA-100, ATA-133, SATA, & SATA2 are marketed to
these users/uses. The enthusiast market is an incredibly fast
growing sector and they bite hard when the lure of specs are thrown
their way. Hardware groups in Usenet & on the web are constantly
overflowing with naive users drooling over and throwing around specs,
even if the posters don't really appreciate the actual
benefits/drawbacks of the new technology. Manufacturers make money by
not being stupid

Is all this a scam? Well I suppose how you answer all depends on how
burnt you feel by the industry. Certainly there is misdirection and
shrewd marketing strategy at work. If you are a sophisticated buyer
it can be easy to adopt a "you get what you deserve" philosophy re
naiveté. But there is no denying articles like the one that inspired
this thread are a part of "Keep the Consumer Dissatisfied" type
marketing and not engineering achievement pushed by users imminent
needs.
 
Back
Top