View Full Version : Ultra high-speed Compact Flash - do you need it?
Here is a summary or excerpt from an article that has just been published on DPNow:
Are the latest generation of ultra high speed UDMA compact flash cards going to speed up your new DSLR? Maybe yes, maybe no. It's a far from straightforward answer, as we found out when testing a Lexar Professional 300x UDMA card and a standard high speed SanDisk Extreme III with three if the latest DSLRs, the fast shooting Canon EOS-40D, Olympus E-3, and Sony Alpha A700. -+- Ref: 4TFD
More... (http://dpnow.com/4445.html)
Here is a summary or excerpt from an article that has just been published on DPNow:
Are the latest generation of ultra high speed UDMA compact flash cards going to speed up your new DSLR? Maybe yes, maybe no. It's a far from straightforward answer, as we found out when testing a Lexar Professional 300x UDMA card and a standard high speed SanDisk Extreme III with three if the latest DSLRs, the fast shooting Canon EOS-40D, Olympus E-3, and Sony Alpha A700. -+- Ref: 4TFD
More... (http://dpnow.com/4445.html)
Well! The Canon EOS doesn't support the latest UDMA high speed memory card technology, but that didn't stop it from out-gunning UDMA-enabled competitors.
Ian
Many thanks for this timely and informative news. I happeded to be at large caemra shops in Tokyo last week and asked for prices for these. I did not buy any as I thought they were still expensive. Now I feel glad that I did not buy it, after reading this news. I did not know Canon does not support UDMA or conversely I misunderstand UDMA can be used on any cameras as long as CF cards are used.
I learned at that time that newer SDHC cards are cheaper than older CF cards for the same memory size. I do not understand why. but it does not matter. My camera does not support SDHC.:rolleyes:
yoshi
Many thanks for this timely and informative news. I happeded to be at large caemra shops in Tokyo last week and asked for prices for these. I did not buy any as I thought they were still expensive. Now I feel glad that I did not buy it, after reading this news. I did not know Canon does not support UDMA or conversely I misunderstand UDMA can be used on any cameras as long as CF cards are used.
I learned at that time that newer SDHC cards are cheaper than older CF cards for the same memory size. I do not understand why. but it does not matter. My camera does not support SDHC.:rolleyes:
yoshi
Hi Yoshi,
Just to clarify things - UDMA compact flash cards are compatible with non-UDMA cameras, but performance may actually be slightly inferior to 'standard' high speed cards rated around 133x.
The market for SD and SDHC is much bigger than Compact Flash so there is economy in the scale of production.
Ian
Stuart B
16-11-07, 12:28 AM
All your "test" has proven is that when shooting a burst it is better to use compressed RAW than uncompressed RAW.
The Canon 40D only has compressed RAW, whereas the A700 has both. You have shot with compressed RAW for the Canon 40D, and uncompressed RAW for the A700, which should not be compared in this fashion.
If you used the same file format for both cameras, then it would be a fair comparison. The A700 can shoot about 30 cRAW frames with a 300X UDMA card, or about 20 frames with a 133X card.
Seems like a rather fishy "article" to me.
Stuart B
16-11-07, 12:31 AM
Well! The Canon EOS doesn't support the latest UDMA high speed memory card technology, but that didn't stop it from out-gunning UDMA-enabled competitors.
Ian
Try repeating the test, except with like-for-like, rather than comparing different file formats for different cameras ...
All your "test" has proven is that when shooting a burst it is better to use compressed RAW than uncompressed RAW.
The Canon 40D only has compressed RAW, whereas the A700 has both. You have shot with compressed RAW for the Canon 40D, and uncompressed RAW for the A700, which should not be compared in this fashion.
If you used the same file format for both cameras, then it would be a fair comparison. The A700 can shoot about 30 cRAW frames with a 300X UDMA card, or about 20 frames with a 133X card.
Seems like a rather fishy "article" to me.
There is nothing 'fishy' at all - assuming your data is correct, I'll of course amend the article.
One wonders why the A700 offers uncompressed RAW in the first place?
The Canon 40D may compress its RAW files, but there is no alternative 'uncompressed' RAW file recording mode.
The Olympus E-3 data will be amended because I later discovered that shooting in JPEG Normal as opposed to JPEG Fine allows the camera to shoot indefinitely at full continuous 5fps burst mode.
Ian
Stuart B
16-11-07, 12:20 PM
There is nothing 'fishy' at all - assuming your data is correct, I'll of course amend the article.
One wonders why the A700 offers uncompressed RAW in the first place?
The Canon 40D may compress its RAW files, but there is no alternative 'uncompressed' RAW file recording mode.
The Olympus E-3 data will be amended because I later discovered that shooting in JPEG Normal as opposed to JPEG Fine allows the camera to shoot indefinitely at full continuous 5fps burst mode.
Ian
Excuse me for my cynicism, and thank you for offering to correct the article.
In terms of buffer-to-memory-cards; the Alpha 700 is a speed king, it is insanely fast, so when I saw an article claiming the Canon 40D was faster I was a little dubious.
As for your question about why the Sony A700 offers both... please see the below FAQ by Sony...
http://support.sony-europe.com/DIME/faq/faq.asp?l=en&c=DIME&sc=DSR&m=DSLR-A700&f=mod_DSC_A700
If you look at the section entitled "Computer, Operating System, and file format?" they have a Q&A called "Are there any differences in picture quality between cRAW and RAW?".
To my knowledge, Canon has no equivalent FAQ answering why it does not offer uncompressed RAW.
Canon 40D's compressed RAW => about 12MB per file (2/3 compression)
Sony A700's compressed RAW => about 12MB per file (2/3 compression)
Sony A700's uncompressed RAW => about 18MB per file (4272 x 2848 x 12 bits \ 8 bits per byte)
If the Canon 40D had uncompressed RAW it would also be about 18MB per file (3888 x 2592 x 14 bits \ 8 bits per byte)
Patrick
16-11-07, 01:23 PM
Excuse me for my cynicism, and thank you for offering to correct the article.
In terms of buffer-to-memory-cards; the Alpha 700 is a speed king, it is insanely fast, so when I saw an article claiming the Canon 40D was faster I was a little dubious.
As for your question about why the Sony A700 offers both... please see the below FAQ by Sony...
http://support.sony-europe.com/DIME/faq/faq.asp?l=en&c=DIME&sc=DSR&m=DSLR-A700&f=mod_DSC_A700
If you look at the section entitled "Computer, Operating System, and file format?" they have a Q&A called "Are there any differences in picture quality between cRAW and RAW?".
To my knowledge, Canon has no equivalent FAQ answering why it does not offer uncompressed RAW.
Canon 40D's compressed RAW => about 12MB per file (2/3 compression)
Sony A700's compressed RAW => about 12MB per file (2/3 compression)
Sony A700's uncompressed RAW => about 18MB per file (4272 x 2848 x 12 bits 8 bits per byte)
If the Canon 40D had uncompressed RAW it would also be about 18MB per file (3888 x 2592 x 14 bits 8 bits per byte)
Your comparisons could also be flawed
The Canon shoots in 14bit, not sure about the Sony do they shoot 12 or 14 bit?
Since the quality difference is claimed to be none between Raw and cRAW the whole question seems to me to be academic for a photographer other than the fact the compressed file will write faster and fit more images on a card.
Patrick
Stuart B
16-11-07, 01:46 PM
Your comparisons could also be flawed
The Canon shoots in 14bit, not sure about the Sony do they shoot 12 or 14 bit?
Since the quality difference is claimed to be none between Raw and cRAW the whole question seems to me to be academic for a photographer other than the fact the compressed file will write faster and fit more images on a card.
Patrick - Sony A700 is 12-bit.
Indeed; my suggestion is that the difference between compressed and uncompressed is academic, except in regards to the speed test which was performed by DPNow. Comparing the A700 shooting uncompressed RAW to the 40D shooting compressed RAW seems a little unbalanced.
Patrick - Sony A700 is 12-bit.
Indeed; my suggestion is that the difference between compressed and uncompressed is academic, except in regards to the speed test which was performed by DPNow. Comparing the A700 shooting uncompressed RAW to the 40D shooting compressed RAW seems a little unbalanced.
I think my example would be very typical of many people new to an A700: 'RAW' is unambiguous. 'cRAW' is something unusual and a lot of people would ignore it and simply use RAW. So I'm still tring to figure out what the benefit of uncompressed RAW could be. My theory is that uncompressed RAW would offer better performanc for very long bursts of continuous shooting than compressed RAW, but that would need to be verified through testing.
Ian
Stuart B
16-11-07, 04:47 PM
An interesting theory but I fail to see your logic. The compressed RAW format is better suited to longer bursts than the uncompressed RAW.
The inclusion of both formats is the same as including different compression options for TIFF files. None of them offer any more detail, yet some of them are smaller than others.
How about showing the results for both compressed and uncompressed RAW on the Sony, rather than picking only one? That way, your readers can make their own minds up which one is best. That would seem a bit more of an impartial and objective way of presenting the true data.
An interesting theory but I fail to see your logic.
How about showing the results for both compressed and uncompressed RAW on the Sony, rather than picking only one?
The camera went back to Sony on Tuesday, unfortunately. They could only let me have it for two weeks.
The logic is quite clear to me - both in choosing, without prior knowledge, between 'RAW' and 'cRAW' and in the likelihood that compress RAW would likely offer a slower continuous shooting rate after the buffer was saturated. This is exactly what happens with the 40D. More in-camera processing retains the bottleneck before the buffer to card interface once the buffer is filled. The A700 does have the benefit of a faster buffer to card interface thanks to UDMA, so my logic is subject to some verification, of course.
I've just noticed your snipe about 'being paid by Canon'. Frankly, if your attitude is like that, you aren't welcome on this site. For the record - I OWN a Sony Alpha A100 and I OWN a Sony Carl Zeiss 16-80 standard zoom. I find the insinuation that my testing is biased quite insulting and the suggestion that Canon paid me to show how much better the 40D is compared to the Olympus E-3 and Sony Alpha A700 - is just laughable. If you are in the UK you will see that the site is currently plastered with Sony Alpha advertising, I run two separate sites aimed at Olympus users, and if only Canon would advertise here (they never have).
If you want to be welcome on this site again, think carefully before displaying your innate paranoia about how the A700 is tested and making completely uncalled for libellous allegations.
Ian
Patrick
16-11-07, 05:33 PM
The camera went back to Sony on Tuesday, unfortunately. They could only let me have it for two weeks.
The logic is quite clear to me - both in choosing, without prior knowledge, between 'RAW' and 'cRAW' and in the likelihood that compress RAW would likely offer a slower continuous shooting rate after the buffer was saturated. This is exactly what happens with the 40D. More in-camera processing retains the bottleneck before the buffer to card interface once the buffer is filled. The A700 does have the benefit of a faster buffer to card interface thanks to UDMA, so my logic is subject to some verification, of course.
I've just noticed your snipe about 'being paid by Canon'. Frankly, if your attitude is like that, you aren't welcome on this site. For the record - I OWN a Sony Alpha A100 and I OWN a Sony Carl Zeiss 16-80 standard zoom. I find the insinuation that my testing is biased quite insulting and the suggestion that Canon paid me to show how much better the 40D is compared to the Olympus E-3 and Sony Alpha A700 - is just laughable. If you are in the UK you will see that the site is currently plastered with Sony Alpha advertising, I run two separate sites aimed at Olympus users, and if only Canon would advertise here (they never have).
If you want to be welcome on this site again, think carefully before displaying your innate paranoia about how the A700 is tested and making completely uncalled for libellous allegations.
Ian
I can't help feeling this quarrel escalating unnecessarily. So what if one camera is a few microseconds faster than another to download it really doesn't matter.
I personally cant comment on the Sony, but for Stuarts B benefit I own 40D and its bloody fast (now there is a good scientific description for you:D ) and I don't doubt the Sony is also bloody fast.
As for uncompressed RAW Sony say there is no difference in quality, why then they have bother with it I can't imagine.
So come on lads calm down and go take a few pictures, much fare more important than download speeds.
Patrick
Stuart_B
16-11-07, 05:59 PM
The camera went back to Sony on Tuesday, unfortunately. They could only let me have it for two weeks.
You are currently showcasing an article claiming to be cutting edge and informative, backed up by impartial evidence, which says that the 40D bursts for longer than the A700.
You have been given quite reasonable feedback, pointing out that your tests were flawed, and as a consequence biased against the A700, enough such as to invalidate your test results and the conclusions you have made based upon those results.
Yet; despite that feedback, you are continuing to boast the same results and conclusion.
My suggestion to you is; either correct your article, or remove it.
The logic is quite clear to me - both in choosing, without prior knowledge, between 'RAW' and 'cRAW' and in the likelihood that compress RAW would likely offer a slower continuous shooting rate after the buffer was saturated.
An interesting theory indeed but where is your evidence?
This is exactly what happens with the 40D. More in-camera processing retains the bottleneck before the buffer to card interface once the buffer is filled. The A700 does have the benefit of a faster buffer to card interface thanks to UDMA, so my logic is subject to some verification, of course.
The A700 (and A100) have faster buffer-to-card speeds even without UDMA. E.g. with a 133x memory card; the A700 manages buffer-to-memory-card speeds of 15MB/sec with a memory card, whereas the 40D only manages 7MB/sec with the same card. With compressed RAW, the A700 can shoot a little over 1fps, unlimited.
I've just noticed your snipe about 'being paid by Canon'. Frankly, if your attitude is like that, you aren't welcome on this site.
...
...
...
If you want to be welcome on this site again, think carefully before displaying your innate paranoia about how the A700 is tested and making completely uncalled for libellous allegations.
As far as I recall, I have not made any such allegations. I did consider making a witty comment to that effect (in jest) but I decided that you might not take my humour very well.
However, I apologise wholeheartedly for any insinuation I may have made. I accept completely that you have made an honest mistake.
Please now correct that mistake.
Stuart_B
16-11-07, 06:23 PM
I can't help feeling this quarrel escalating unnecessarily. So what if one camera is a few microseconds faster than another to download it really doesn't matter.
I personally cant comment on the Sony, but for Stuarts B benefit I own 40D and its bloody fast (now there is a good scientific description for you:D ) and I don't doubt the Sony is also bloody fast.
As for uncompressed RAW Sony say there is no difference in quality, why then they have bother with it I can't imagine.
So come on lads calm down and go take a few pictures, much fare more important than download speeds.
Patrick
If the accuracy of the results and validity of the conclusion of the article are so irrelevant, then what was the point of the article in the first place?
1. both cameras measure up pretty equally with 133x memory cards; the 40D has a larger buffer, whereas the A700 has a faster buffer-to-card speed
2. the A700 is significantly faster with 300x memory cards; the 40D does not improve much, whilst the A700 improves a great deal
E.g. A700 can take 30 cRAW frames at 5fps, with a Lexar 300X UDMA card, which is about one-and-a-half-times as many as the Canon 40D can take at 6.5fps.
If the accuracy of the results and validity of the conclusion of the article are so irrelevant, then what was the point of the article in the first place?
1. both cameras measure up pretty equally with 133x memory cards; the 40D has a larger buffer, whereas the A700 has a faster buffer-to-card speed
2. the A700 is significantly faster with 300x memory cards; the 40D does not improve much, whilst the A700 improves a great deal
E.g. A700 can take 30 cRAW frames at 5fps, with a Lexar 300X UDMA card, which is about one-and-a-half-times as many as the Canon 40D can take at 6.5fps.
I made considered amendments to the article (on the test results page) earlier in the day - maybe you hadn't noticed this?
I can't publish test data for cRAW because I haven't done the test and I don't have an A700 any more to do that test. If I did, I'd have run the test immediately - but I have promised to do the test when Sony can provide me with another camera.
In the mean time, the article is honest and as clear as it can be in the circumstances. It is not misleading in any way either. I am not going to take it down.
What I feel has been highlighted is that there probably doesn't appear to be any requirement for the standard uncompressed RAW mode. The Olympus E-3 uses a compressed RAW mode by default but Olympus feels no need to call it 'compressed RAW' and offer an uncompressed alternative. In which case, Sony is inviting its users to reduce the performance of their camera for no apprently good reason.
Anyway, I look forward verifying your data as soon as I get the opportunity.
Ian
Patrick
16-11-07, 06:50 PM
If the accuracy of the results and validity of the conclusion of the article are so irrelevant, then what was the point of the article in the first place?
1. both cameras measure up pretty equally with 133x memory cards; the 40D has a larger buffer, whereas the A700 has a faster buffer-to-card speed
2. the A700 is significantly faster with 300x memory cards; the 40D does not improve much, whilst the A700 improves a great deal
E.g. A700 can take 30 cRAW frames at 5fps, with a Lexar 300X UDMA card, which is about one-and-a-half-times as many as the Canon 40D can take at 6.5fps.
If and I repeat if its inaccurate, Ian wouldn't be the first nor will he be the last to make a mistake. As to the niceties of the figures claimed by the manufacturers, these are often inaccurate in practical use as they cherry pick the way they are quoted.
As matter of interest all the reports I have read on the 40D state it runs faster than Canon claim until the buffer is almost filled then it slows down.
Personally and I bet this applies to most readers I couldn't give a S*** which is the faster, they are both good cameras giving top class results, the rest is for anoraks.
Patrick
If and I repeat if its inaccurate, Ian wouldn't be the first nor will he be the last to make a mistake. As to the niceties of the figures claimed by the manufacturers, these are often inaccurate in practical use as they cherry pick the way they are quoted.
As matter of interest all the reports I have read on the 40D state it runs faster than Canon claim until the buffer is almost filled then it slows down.
Personally and I bet this applies to most readers I couldn't give a S*** which is the faster, they are both good cameras giving top class results, the rest is for anoraks.
Patrick
I agree entirely - but just for the record, I haven't made a mistake. Every statement in the article is technically correct. All that's been brought to light is that if you use the camera differently, the results may be different.
I don't see any need to add to this thread until I get to do additional tests! Thanks for your interest!
Ian
Unregistered
16-11-07, 10:36 PM
I agree entirely - but just for the record, I haven't made a mistake. Every statement in the article is technically correct. All that's been brought to light is that if you use the camera differently, the results may be different.
I don't see any need to add to this thread until I get to do additional tests! Thanks for your interest!
Ian
Except the Sony doing 12 MP images does them endlessly in full speed and you compared a 10 MP compressed RAW with an 12 MP uncompressed raw then state the camera that all your testing methodoligy favored impressed you.
Doing fair tests is takes some research..
Stuart_B
16-11-07, 10:43 PM
Personally and I bet this applies to most readers I couldn't give a S*** which is the faster, they are both good cameras giving top class results, the rest is for anoraks.
Well; I guess then that Sandisk, Sony, Nikon, Olympus, Lexar, and a lot of other manufacturers have made a serious goof themselves - because they are trying to market high-speed memory when nobody (except for anoraks) really wants it.
Oh well!
Thankfully, it seems like high speed memory cards sell like gold dust. So there must be an awful lot of anoraks out there. So many anoraks out there, in fact, that Dpnow has writen this very article, about an issue which nobody else really cares about.
Patrick
16-11-07, 11:04 PM
Well; I guess then that Sandisk, Sony, Nikon, Olympus, Lexar, and a lot of other manufacturers have made a serious goof themselves - because they are trying to market high-speed memory when nobody (except for anoraks) really wants it.
Oh well!
Thankfully, it seems like high speed memory cards sell like gold dust. So there must be an awful lot of anoraks out there. So many anoraks out there, in fact, that Dpnow has writen this very article, about an issue which nobody else really cares about.
Read what I said, I didn't care which was faster not that cameras in general shouldn't be faster.
Its the difference between the two that is for anoraks.
Cameras are for making pictures not for fussing over which is a micro second faster than another.
Patrick
Read what I said, I didn't care which was faster not that cameras in general shouldn't be faster.
Its the difference between the two that is for anoraks.
Cameras are for making pictures not for fussing over which is a micro second faster than another.
Patrick
One of the reasons I produced the article was to inform people that they didn't necessarily have to go for the latest 300x cards and so they could save a fair bit of money. At the same time, people whose cameras could benefit from UDMA technology could be happy that their choice of investing in such cards may be of benefit when shooting RAW.
Saving money or getting value for money is always important in my book :)
Ian
Bearface
17-11-07, 10:08 AM
Well; I guess then that Sandisk, Sony, Nikon, Olympus, Lexar, and a lot of other manufacturers have made a serious goof themselves - because they are trying to market high-speed memory when nobody (except for anoraks) really wants it.
Oh well!
Oh well! :rolleyes:
Are these the same kind of anoraks who join up to photography forums and immediately employ extreme pedantry to split hairs over an in-house article on memory cards, instead of participating positively to the site and perhaps even benefitting from it..?
Some people have just got too much time on their hands, eh Stuart? And as for caring whether or not the Sony A700 has a faster buffer than the Canon 40D.........well, obviously that's really important to technoge.....sorry, I mean those of us who are actually into serious photography.
Still, I'm sure these anoraks already have a place to argue (incessantly...) over the finer points of er.......camera's buffer speeds and all that nonsense.......a place where the pedants can take delight in correcting or criticizing semantics or what they perceive to be fundamental errors in other people's postings. Still, I'll bet it's a very worthy and fulfilling pastime...
Sadly I doubt such folk have much time left over for that joyous pursuit that I like to call "photography". :D
Patrick
17-11-07, 10:27 AM
Oh well! :rolleyes:
Are these the same kind of anoraks who join up to photography forums and immediately employ extreme pedantry to split hairs over an in-house article on memory cards, instead of participating positively to the site and perhaps even benefitting from it..?
Some people have just got too much time on their hands, eh Stuart? And as for caring whether or not the Sony A700 has a faster buffer than the Canon 40D.........well, obviously that's really important to technoge.....sorry, I mean those of us who are actually into serious photography.
Still, I'm sure these anoraks already have a place to argue (incessantly...) over the finer points of er.......camera's buffer speeds and all that nonsense.......a place where the pedants can take delight in correcting or criticizing semantics or what they perceive to be fundamental errors in other people's postings. Sadly I doubt such folk have much time left over for that joyous pursuit that I like to call "photography". :D
Well said:D
Patrick
Stuart_B
17-11-07, 11:30 PM
One of the reasons I produced the article was to inform people that they didn't necessarily have to go for the latest 300x cards and so they could save a fair bit of money. At the same time, people whose cameras could benefit from UDMA technology could be happy that their choice of investing in such cards may be of benefit when shooting RAW.
Saving money or getting value for money is always important in my book :)
Ian
I am sorry then, I must have completely mistook your meaning before, when you said...
Well! The Canon EOS doesn't support the latest UDMA high speed memory card technology, but that didn't stop it from out-gunning UDMA-enabled competitors.
Ian
... because from that it sounded like there was a competition to be had, and you had declared Canon to be the winner.
Unregistered
17-11-07, 11:35 PM
Oh well! :rolleyes:
Are these the same kind of anoraks who join up to photography forums and immediately employ extreme pedantry to split hairs over an in-house article on memory cards, instead of participating positively to the site and perhaps even benefitting from it..?
Some people have just got too much time on their hands, eh Stuart? And as for caring whether or not the Sony A700 has a faster buffer than the Canon 40D.........well, obviously that's really important to technoge.....sorry, I mean those of us who are actually into serious photography.
Still, I'm sure these anoraks already have a place to argue (incessantly...) over the finer points of er.......camera's buffer speeds and all that nonsense.......a place where the pedants can take delight in correcting or criticizing semantics or what they perceive to be fundamental errors in other people's postings. Still, I'll bet it's a very worthy and fulfilling pastime...
Sadly I doubt such folk have much time left over for that joyous pursuit that I like to call "photography". :D
Talking about lame insults
They might just be the type who spend their hours shooting frames at blank walls so that they can say one camera is faster than another camera... Or, the type who actually spend their days reading about other people who spend hours shooting at blank walls...
Zzzzzzz...
Are you done being childish yet?
Bearface
18-11-07, 12:06 AM
Talking about lame insults
They might just be the type who spend their hours shooting frames at blank walls so that they can say one camera is faster than another camera... Or, the type who actually spend their days reading about other people who spend hours shooting at blank walls...
Zzzzzzz...
Are you done being childish yet?
Well it certainly prompted an angry response from you, didn't it..?
Sorry "unregistered", I'm only able to enter into a battle of wits with individuals who are armed with some in the first place. Plus, I refuse to engage with anyone who ducks registration and posts anonymously...
Hello,
I would like to see if I can add to this in a way that clears up a bit of the confusion. I am a Nikon user, and have owned a D100, a D200, and now have a D300. For both Nikon and Canon, who have been making DSLR's with large image file sizes and raw modes for a while, the reason for a lot of this complexity is two-fold, IMHO.
First, the reason for having the option of compressed vs. uncompressed raw files has a lot to do with the limits of the early cameras and cards. Raw file formats, with a few exceptions, are proprietary to the camera maker, and the specific version of the Nikon Electronc Format (NEF) raw file used by the D100, and I am pretty certain the D200 as well, did lose just a small amount of quality when compressed, because the compression was very slightly lossy. The compressed file size was about half the uncompressed size, and this was a big deal when a 1 GB CF card cost $200. At the same time, the buffer was much smaller by today's standards, and the camera's electronics were slower.
In addition, the D100 could also be used tethered to a computer with lots of storage, and did not have to have a card installed for this.
In this context, it made sense to give the user the option of compressed or uncompressed raw, so they could decide which compromise hurt the least- to shoot uncompressed, which was faster on the D100 because the compression was the bottleneck, not the card write speed, or shoot compressed, which was slower but got twice as many shots on a card (not an issue with tethered operation).
Even back then (2002?), there was a narrowly adopted standard called Write Accelleration supported by some cameras (including the D100) and some vendors (I think Lexar started it) but it never really took off. It was an early attempt to do the same thing as UDMA.
The other reason for a choice between compressed and uncompressed raw formats is that shooting raw means using raw conversion software, and if your favorite raw converter did not support the compressed format, your only option was to use uncompressed raw.
Newer cameras have been able to bypass all of this legacy, and for example have only one raw format, with no choice for the user between compressed or uncompressed. The raw conversion software adopts this one format from the outset.
The D300 and D3X even go so far as to let the user choose between 12 bit NEF and 14 bit NEF through the menu, but this is really more of a pro-level feature.
As the pixel counts have gone up, the image sizes have gotten bigger. The image processors have gotten faster (and Canon sometimes boost performance by using more than one of them), the buffers bigger, and the card interface faster. The high end cameras also have much higher maximum frame rates, with 6 to 10 fps now common.
This is why hands-on testing is so helpful. For a given UDMA compatible camera, configured a certain way, shooting at a certain rate, using a UDMA card may or may not help, because it may or may not be the bottleneck. An ordinary card may be more than fast enough shooting JPEG at a moderate rate. UDMA is more likely to help when a fully compatible camera with a high pixel count shoots uncompressed raw at 8 or 10 fps for a long burst. Remove any of these factors and a high quality non-UDMA card may be just fine.
All this still leaves me wondering how fast any card has to be before it will not slow down my D300 shooting 14 bit uncompressed NEF's at 6 fps. :)
On the software side, only the very most recent update of Lightroom, for example, even supports this camera.