SD/CF card speeds - A marketing gimmick?

Computer Hardware Memory Cards & USB Sticks
#1

I am new to photography and using memory for cameras. When I was planning to buy SD memory, I saw so many choices and did some studies and the results I got were very surprising to me.

So in essence, the speed of SD/CF cards matter ONLY when you’re reading from them, not for performance with the cameras.

http://www.kenrockwell.com/tech/cf-sd.htm
http://www.bobatkins.com/photography/digital/compact_flash_memory_cards.html

See the second link - Most cameras, even high end cannot write faster than 1.4 MB/s - so no point in buying the faster memory. And all cameras have sufficient internal buffer which is difficult to run out unless shooting RAW sports mode.

I think I made a wise decision buying a 2 GB Kingston SD for my Canon SD800 IS from Frys for $25 (Thanks ghettocowboy for letting me know the bargain) and not the $90 or so ‘Ultra speed’ elsewhere

Another gimmick forcing unknowing users to invest $$$ in something they dont need.

#2

I suspect the faster write speeds, excelling the speed that of the camera read ability is for people who have a dedicated slot and like to save the precious few seconds when it comes to data transfer of those pics onto their PC’s.

Also prob not a bad thing to have for possible future camera products that might take advantage of the higher read/write speeds.

I would buy it but only if the diff in price between them would be less than $10.

#3

the random access time will increase the rate at which you can shot pictures i belive

#4

The ‘Ultra’ speed cards were about twice the cost of the standard ones when I bought it yesterday. So I cannot justify paying $50 for a card that costs $25

As for future cameras, I will not retain my $25 card when I buy my new one in a few years. After all the new camera will have more features, higher file (megapixel) size which means more storage needed, so I will probably upgrade my storage card too and buy a better (higher storage) one.

Using a faster card for cameras seems like using USB 2.0 instead of 1.1 for using a mouse.

#5

difference is 50bux… whew ya I too would not go for it

#6

For the first sentence to be valid, it also depends on the card itself. Many manufacturers rate their speeds by their read speed, ignoring the fact that their write speed may be far slower than their read speed even under optimal benchmarking conditions. In fact, I’ve tested non-speed-rated cards (e.g. a “Made In Japan” PNY 2GB SD card with no speed rating) which have higher read speeds than some speed-rated cards (60X SD cards from both Lexar and Sandisk that I’ve tested); however, the PNY can muster only slightly faster write speed than the plain “unrated” Sandisk SD card in benchmarks (both “unrated” cards can write at between 33X and 40X in my testing).

As for the second sentence, this is incorrect. A high-end digital SLR can write at up to about 8 MB/s. And many of the newer Canon cameras (both SLR and point-and-shoot) with the Digic II processing chip can write at 5 MB/s or faster. However, these cameras can take advantage of such higher-speed cards only under the most-stressful scenario: continuous rapid-sequence shooting at the cameras’ maximum resolution and image quality settings and high ISO sensitivities.

In other words, the higher end your camera, the more likely you’ll need a higher-speed card.

On the other hand, if the faster card is significantly more expensive than a “slower” non-rated card, then the faster card is not worth the extra cost, IMHO.

#7

I did a further test on a modern low-end digital SLR camera (Nikon D50), and found that the 60x speed-rated (this refers to the write speed, not the read speed) cards are slightly faster than their non-speed-rated brandmates, but not by as much as the memory card manufacturers lead you to believe. With continuous shooting of six compressed RAW images, the 60x cards (from Lexar and Sandisk) took about 10 to 11 seconds (from first shutter release to the completion of the writing), while the slowest non-speed-rated cards (of those that I have on hand) took nearly 16 seconds. Two blue non-speed-rated SanDisk SD cards proved to be either “fast” or “slow,” depending on the capacity of the card (the 1GB card, with a RAW shooting performance time of about 12 seconds, performed more like the 60x cards in my testing, while the 256MB card is more like the “slow” PNY 2GB card). What’s more, the results show very little correlation to the TRT results with a card reader connected to a high-performance PC.

Please note that the above performance results are relative. The 60x SD cards are the fastest SD cards that are available from most big-box retailers like Wal-Mart, Best Buy and Circuit City. You’ll pretty much have to buy 133x-rated SD cards from a large camera store or an online reseller.

#8

Read and/or write performance may also depend on the circuit that interfaces with the SD card, and unless you know for certain that it’s not the culprit, you can’t be sure what you measuring is the throughput of the card, or the circuit.

#9

For all practical purposes, I agree. In addition, you may notice an increase in write performance with certain cards in a given camera because the camera’s circuit works better with those particular cards, not because of whether that card is of high-speed or low-speed.

And I have checked the throughput of my particular card reader: It has a higher throughput than all of the SD cards in my possession. The fastest-writing card in my possession (a 60x Lexar Platinum II) tested out to have a write speed that exceeds the card manufacturer’s 60x rating - about 83x.

That said, some ultra-high-megapixel cameras really need ultra-high-speed cards (133x or faster) in order to function properly.

#10

One more thing about SD card speed:

Eventually, all SD cards manufactured beginning this year will have a speed rating; non-speed-rated SD cards and those with arbitrary speed ratings will become a thing of the past. There will be three speed classes of SD cards - SD Class 2, SD Class 4 and SD Class 6 - each corresponding to the guaranteed minimum transfer rate (in MB/s). For example, the current Toshiba standard SD cards (and also those manufactured by Toshiba for PNY) are graded SD Class 4, which means the cards have a guaranteed minimum transfer rate of 4 MB/s. SanDisk has three lines of SD cards, each with a different speed rating: The regular SanDisk cards are rated SD Class 2 (2 MB/s); the Ultra II, SD Class 4 (even though the Ultra II was bench-tested at 60x write/66x read); the Extreme III, SD Class 6 (6 MB/s). Lexar currently has no SD Class 4-rated cards; its regular line is SD Class 2-rated, while its current Platinum II 60x card line is SD Class 6-rated. The existing and other speed ratings, if used, will supplement the SD class ratings.

The bad thing? Because of this, some brands of SD Class 2 cards will sell for as much as some brands of SD Class 4 cards, while some brands of SD Class 6 cards will sell for the same price as or a lower price than some brands of SD Class 4 cards; for example, Lexar’s consumer SD Class 6 card currently sells for the same price as SanDisk’s SD Class 4 card.

#11

[QUOTE=RJL65;1633631]As for the second sentence, this is incorrect. A high-end digital SLR can write at up to about 8 MB/s. And many of the newer Canon cameras (both SLR and point-and-shoot) with the Digic II processing chip can write at 5 MB/s or faster. However, these cameras can take advantage of such higher-speed cards only under the most-stressful scenario: continuous rapid-sequence shooting at the cameras’ maximum resolution and image quality settings and high ISO sensitivities.

In other words, the higher end your camera, the more likely you’ll need a higher-speed card.

On the other hand, if the faster card is significantly more expensive than a “slower” non-rated card, then the faster card is not worth the extra cost, IMHO.[/QUOTE]

That second sentence has now been updated - again. The newest DSLRs, such as the Canon 40D, 1D Mk III and 1Ds Mk III, and the Nikon D300 and D3, can write to many high-speed cards about as fast as the cards themselves can go. In fact, the 1Ds Mk III, the D300 and the D3 can write to the fastest CF cards at well over 20 MB/s (133x). For those three cameras, it does make sense to buy the very fastest CF cards if you’re shooting a lot of action shots which require the use of the sequential (continuous) shooting mode, especially with the highest-quality JPEGs or RAW images. Currently, the Canon 1D Mk III and the Canon 1Ds Mk III are the only DSLRs which can take full advantage of any SD or SDHC card with a speed rating faster than 60x.

#12

Some of the speeds are overstated. The only way to get a really good idea of some cards’ performance is to go to stevesforums (there is a main page dedicated to camera reviews and it’s the best site I’ve found overall on reviewing cameras and they do give you ideas of SD and other card performances with the cameras they test; oddly enough, about the slowest/smallest they use consistently is 50x 2GB from Kingston (Elite or Elite Pro) card.

I was kind of surprised to not find in all the replies that the Sandisk Extreme III line of cards are the best-performing cards in all categories, as proven by at least one study in the last 3-4 months. If one needs the SD/SDHC or other memory when they have their rebate promotions on and can afford the initial outlay of $$$ and wait 8 weeks for the “rebite” (as I love to call them), you’ll get (recently) 3 8gb Class 6 Extreme III cards for about $15 each. I’ve proven to myself by using SiSoft Sandra Pro to benchmark my Sandisk Extreme III 4gb SDHC cards are indeed significantly faster than my Kingston 8gb Class 6 cards by a noticeable margin. The Sandisk (whether Ultra II or Extreme III) cards are the closest cards I have to match their rated speeds. Here’s the data from one testing session between the Sandisk Extreme III 4gb cards and Kingston 8gb Class 6:

SiSoftware Sandra

Benchmark Results
Device Index : 3002ops/minute
Endurance Factor : 37.20
Results Interpretation : Higher index values are better.

Performance vs. Speed
Drive Index : 6.25ops/minute/Mbps
Results Interpretation : Higher index values are better.

512bytes Files Test
Read Performance : 25570ops/minute, 213.08kB/s, 1.42X
Write Performance : 1725ops/minute, 14.38kB/s, 0.10X
Delete Performance : 2563ops/minute
File Fragments : 1.0
Combined Index : 3886ops/minute

32kB Files Test
Read Performance : 14774ops/minute, 7.69MB/s, 52.53X
Write Performance : 1923ops/minute, 1.00MB/s, 6.84X
Delete Performance : 2694ops/minute
File Fragments : 1.0
Combined Index : 3786ops/minute

256kB Files Test
Read Performance : 3806ops/minute, 15.86MB/s, 108.26X
Write Performance : 1243ops/minute, 5.18MB/s, 35.36X
Delete Performance : 2773ops/minute
File Fragments : 1.0
Combined Index : 1910ops/minute

2MB Files Test
Read Performance : 541ops/minute, 18.03MB/s, 123.11X
Write Performance : 258ops/minute, 8.60MB/s, 58.71X
Delete Performance : 2076ops/minute
File Fragments : 1.0
Combined Index : 346ops/minute

64MB Files Test
Read Performance : 17ops/minute, 18.13MB/s, 123.79X
Write Performance : 12ops/minute, 12.80MB/s, 87.38X
Delete Performance : 1223ops/minute
File Fragments : 1.0
Combined Index : 13ops/minute

256MB Files Test
Read Performance : 4ops/minute, 17.07MB/s, 116.51X
Write Performance : 3ops/minute, 12.80MB/s, 87.38X
Delete Performance : 543ops/minute
File Fragments : 1.0
Combined Index : 3ops/minute

Performance Test Status
Run ID : SanDisk SDDR-113 4GB (USB)
Platform Compliance : x86
System Timer : 3.58MHz

Endurance Test Status
Operating System Disk Cache Used : No
Use Overlapped I/O : No
Test File Size : 32MB
Block Size : 512bytes
File Fragments : 1

Endurance Benchmark Breakdown
Repeated Sector ReWrite : 296kB/s
Sequential Sector Write : 257kB/s
Random Sector Write : 5kB/s

Drive
Total Size : 3.79GB
Free Space : 3.51GB, 93%
Cluster Size : 32kB

Physical Disk
Manufacturer : SanDisk (Extreme III 4gb Class 6 card)
Model : SDDR-113
Version : 9412
Interface : USB
Removable Drive : Yes
Queueing On : No

Performance Tips
Notice 5901 : CD 1x= 150kB/s; Exercise caution when comparing measured versus published CD ratings.
Notice 5008 : To change benchmarks, click Options.
Notice 5004 : Synthetic benchmark. May not tally with ‘real-life’ performance.
Notice 5006 : Only compare the results with ones obtained using the same version!
Notice 5207 : Consider using the File System Benchmark for non-Flash devices.
Notice 5900 : Endurance factor can only be used on the same type of device (SLC or MLC).
Tip 11 : Use the ‘Switch Chart Type’ button to switch between Detailed and Combined charts.
Tip 2 : Double-click tip or press Enter while a tip is selected for more information about the tip.

SiSoftware Sandra

Benchmark Results
Device Index : 2741ops/minute
Endurance Factor : 31.00
Results Interpretation : Higher index values are better.

Performance vs. Speed
Drive Index : 5.71ops/minute/Mbps
Results Interpretation : Higher index values are better.

512bytes Files Test
Read Performance : 25653ops/minute, 213.78kB/s, 1.43X
Write Performance : 1591ops/minute, 13.26kB/s, 0.09X
Delete Performance : 2563ops/minute
File Fragments : 1.0
Combined Index : 3686ops/minute

32kB Files Test
Read Performance : 14912ops/minute, 7.77MB/s, 53.02X
Write Performance : 1663ops/minute, 886.93kB/s, 5.91X
Delete Performance : 2472ops/minute
File Fragments : 1.0
Combined Index : 3448ops/minute

256kB Files Test
Read Performance : 3775ops/minute, 15.73MB/s, 107.38X
Write Performance : 966ops/minute, 4.02MB/s, 27.48X
Delete Performance : 2692ops/minute
File Fragments : 1.0
Combined Index : 1679ops/minute

2MB Files Test
Read Performance : 541ops/minute, 18.03MB/s, 123.11X
Write Performance : 191ops/minute, 6.37MB/s, 43.46X
Delete Performance : 2382ops/minute
File Fragments : 1.0
Combined Index : 305ops/minute

64MB Files Test
Read Performance : 17ops/minute, 18.13MB/s, 123.79X
Write Performance : 7ops/minute, 7.47MB/s, 50.97X
Delete Performance : 1612ops/minute
File Fragments : 1.0
Combined Index : 11ops/minute

256MB Files Test
Read Performance : 4ops/minute, 17.07MB/s, 116.51X
Write Performance : 2ops/minute, 8.53MB/s, 58.25X
Delete Performance : 584ops/minute
File Fragments : 1.0
Combined Index : 3ops/minute

Performance Test Status
Run ID : SanDisk SDDR-113 8GB (USB) [Kingston 8gb Class 6 SDHC]
Platform Compliance : x86
System Timer : 3.58MHz

Endurance Test Status
Operating System Disk Cache Used : No
Use Overlapped I/O : No
Test File Size : 32MB
Block Size : 512bytes
File Fragments : 1

Endurance Benchmark Breakdown
Repeated Sector ReWrite : 197kB/s
Sequential Sector Write : 81kB/s
Random Sector Write : 3kB/s

Drive
Total Size : 7.47GB
Free Space : 7.47GB, 100%
Cluster Size : 32kB

Physical Disk
Manufacturer : SanDisk
Model : SDDR-113
Version : 9412
Interface : USB
Removable Drive : Yes
Queueing On : No

Performance Tips
Notice 5901 : CD 1x= 150kB/s; Exercise caution when comparing measured versus published CD ratings.
Notice 5008 : To change benchmarks, click Options.
Notice 5004 : Synthetic benchmark. May not tally with ‘real-life’ performance.
Notice 5006 : Only compare the results with ones obtained using the same version!
Notice 5207 : Consider using the File System Benchmark for non-Flash devices.
Notice 5900 : Endurance factor can only be used on the same type of device (SLC or MLC).
Tip 11 : Use the ‘Switch Chart Type’ button to switch between Detailed and Combined charts.
Tip 2 : Double-click tip or press Enter while a tip is selected for more information about the tip.

IF you need the speed or at least want to be sure you’re getting it, use the Extreme III line when you can catch the rebate promotions at Adorama; if you just need the storage capacity without having to change cards often, get the Kingston 8gb Class 6 card. You can probably find the best price comparisons from www.pricespider.com (1st choice) or www.pricegrabber.com (2nd choice, and doesn’t list as many vendors). :wink:

#13

Those cards mentioned above seem alright. But if you don’t need 8Gb you can get some pretty good deals these days. I’ve always been random with my SD card purchases. For some reason they fascinate me.

I’ve played with many 3 party SD cards. It surprises me that overall they’re mostly the same in a compact camera. Only in the professional models or for movie filming do you really require a 133x card.

IMPACT cards are one of the more interesting finds. They’re really re-branded Lexar chips. (well 512 [4 for 35$]from China and 1Gb [14$] from Korea) I haven’t seen the others.

Canon’s free sd cards are Japanese made Lexars.

Dane-elec cards are very akward. I bought a 3 pack of 1Gb chips for 25$ in store. All three were from different countries: Korea, China and Taiwan. Each of which performed slightly differently, but overall signiffigantly better than lexar’s. They have SMI chips.

Also Two Taiwanese 2GB Lexars (2 for 40$, they are the pits. unless you’re backing up things for storage I’d never buy those again.

#14

[QUOTE=Matthew12222;2162477]Those cards mentioned above seem alright. But if you don’t need 8Gb you can get some pretty good deals these days. I’ve always been random with my SD card purchases. For some reason they fascinate me.

I’ve played with many 3 party SD cards. It surprises me that overall they’re mostly the same in a compact camera. Only in the professional models or for movie filming do you really require a 133x card.

IMPACT cards are one of the more interesting finds. They’re really re-branded Lexar chips. (well 512 [4 for 35$]from China and 1Gb [14$] from Korea) I haven’t seen the others.

Canon’s free sd cards are Japanese made Lexars.

Dane-elec cards are very akward. I bought a 3 pack of 1Gb chips for 25$ in store. All three were from different countries: Korea, China and Taiwan. Each of which performed slightly differently, but overall signiffigantly better than lexar’s. They have SMI chips.

Also Two Taiwanese 2GB Lexars (2 for 40$, they are the pits. unless you’re backing up things for storage I’d never buy those again.[/QUOTE]

I agree that the ultra-high-speed cards are overkill for most compact point-and-shoot cameras. However, if you have a 10 MP+ digital SLR camera (and I mean a true interchangeable-lens SLR, not a superzoom compact which looks like an SLR), then the 133x+ cards would be worth considering, especially for continuous-burst action shots.

And Lexar’s standard SD/SDHC cards are all over the map, as far as their performance goes. The ones made in Korea perform significantly better than those made in China, which in turn perform somewhat better than those made in Taiwan. In fact, I have a Korean-made 4GB SDHC Class 2 Lexar which has a faster write speed than a Japanese-made PNY-branded Toshiba 4GB SDHC Class 4 card (the read speed on both cards are about equal). And that Korean Lexar Class 2 SDHC card is nearly as fast in real-world performance as many of the Class 6 SDHC cards, even in a higher-level DSLR which takes SDHC cards (such as the Nikon D90 and the Canon EOS 1Ds Mark III).

#15

[QUOTE=RJL65;2163049]I agree that the ultra-high-speed cards are overkill for most compact point-and-shoot cameras. However, if you have a 10 MP+ digital SLR camera (and I mean a true interchangeable-lens SLR, not a superzoom compact which looks like an SLR), then the 133x+ cards would be worth considering, especially for continuous-burst action shots.

And Lexar’s standard SD/SDHC cards are all over the map, as far as their performance goes. The ones made in Korea perform significantly better than those made in China, which in turn perform somewhat better than those made in Taiwan. In fact, I have a Korean-made 4GB SDHC Class 2 Lexar which has a faster write speed than a Japanese-made PNY-branded Toshiba 4GB SDHC Class 4 card (the read speed on both cards are about equal). And that Korean Lexar Class 2 SDHC card is nearly as fast in real-world performance as many of the Class 6 SDHC cards, even in a higher-level DSLR which takes SDHC cards (such as the Nikon D90 and the Canon EOS 1Ds Mark III).[/QUOTE]

I was only partially correct on the Lexar 4GB Class 2 SDHC card. My faster results with that card are based on tests in a high-speed internal card reader. Now that I have a Nikon D90 to test out, it tested out to be significantly slower in-camera than it was in the card reader. Despite the 10MB/sec benchmarked write speed of that card, it wrote at about half that speed - about 5MB/sec - in the D90. Most of the other SD and SDHC cards I have tested perform much closer to their card-reader-benchmarked speeds in the D90. And the D90 can take full advantage of many of the 133x SD/SDHC cards on the market (for example, it tested out at nearly 17MB/sec with a Lexar Professional 133x card - an excellent write-speed result for that card in that camera, given the writing of multiple smaller files).

I have not yet tried the new SanDisk Extreme III 30 MB/s Edition (which is available only in the SDHC sizes from 4GB to 16GB; a 2GB SD version of this edition is not offered). But given the excellent in-camera write speed results of the D90, I suspect that the D90 would take at least some advantage of the new ultra-high-speed cards. And SanDisk’s high-speed cards are among the fastest in their SD speed classes; not surprisingly, the 4GB “Class 4” SDHC Ultra II 15 MB/s Edition (which tested out to about 13MB/sec) outperformed both of the “Class 6” SDHC cards from A-Data and ATP that I had on hand (which tested out to only about 11MB/sec).

#16

I am still curious why ScanDisk would rate themselves as a class 4 card. When it’s obvious that it beats standard class 6’s. Also I’ve seen class 6s such as a verbatim 4GB premium (~9.7mbps) that was whipped by my hp and dane-elec 8GB’s (~11.4mbps), both class 4’s. The difference was noticeable in my brothers Kodak SLR.

I’ve noted that Japaness and Koeran cards are better performers. Over Chinese and Twianese cards. May seem odd, but I’ve seen same brand/ same model but made in different countries and it makes a difference.

#17

[QUOTE=Matthew12222;2178092]I’ve noted that Japaness and Koeran cards are better performers. Over Chinese and Twianese cards. May seem odd, but I’ve seen same brand/ same model but made in different countries and it makes a difference.[/QUOTE]

Very true for the Samsung-made cards (the Korean-made cards are by far better performers than the Chinese- or Taiwanese-made cards) and the Toshiba-made cards (the Japanese-made cards are more reliable than the Taiwanese-made ones). Though both companies market their cards under their own brand names, you’re much more likely to find those cards under other companies’ brand names (Lexar comes to mind for the Samsung-based cards; PNY, Kingston and Dane-Elec are some of the major distributors of Toshiba-based cards).

And not all Chinese-made cards are particularly poor performers: SanDisk (an American-based company) makes all of its cards in China; however, the two higher-speed lines (the current Ultra II 15 MB/s Edition and the Extreme III 30 MB/s Edition) are stellar performers for their classes (the standard blue-colored SanDisk card line is merely average for its class). The older Extreme III (20 MB/s) was also an excellent performer (testing out to about 16MB/sec to 16.5MB/sec in my D90); however, it has been superceded by the new 30 MB/s Edition.

#18

[QUOTE=ExpertTech;1628384]I am new to photography and using memory for cameras. When I was planning to buy SD memory, I saw so many choices and did some studies and the results I got were very surprising to me.

So in essence, the speed of SD/CF cards matter ONLY when you’re reading from them, not for performance with the cameras.

http://www.kenrockwell.com/tech/cf-sd.htm
http://www.bobatkins.com/photography/digital/compact_flash_memory_cards.html

See the second link - Most cameras, even high end cannot write faster than 1.4 MB/s - so no point in buying the faster memory. And all cameras have sufficient internal buffer which is difficult to run out unless shooting RAW sports mode.

I think I made a wise decision buying a 2 GB Kingston SD for my Canon SD800 IS from Frys for $25 (Thanks ghettocowboy for letting me know the bargain) and not the $90 or so ‘Ultra speed’ elsewhere

Another gimmick forcing unknowing users to invest $$$ in something they dont need.[/QUOTE]

Notice that at the time those two articles were first written, the only digital SLR cameras from either Canon or Nikon that was available at the time which used SD cards rather than CF cards was the Canon EOS 1D Mk II and 1Ds Mk II and the Nikon D50. And the D50 maxed out at about 3.5~4 MB/s even with ultra-high-speed cards (the 1D Mk II and the 1DS Mk II, which used CF and/or SD cards, wrote no faster to SD cards than to CF cards with their extant firmware of the time).

The Nikon D40 and D80 had just been introduced at the time of the original post in this thread. However, the D80 could barely take full advantage of even a 60x-rated card while the D40 (which is still in current production) does not write much faster than the D50 did.

Since then, times have certainly changed. The newest CF-based DSLRs can certainly take full advantage of most of the ultra-high-speed cards (with cameras such as the Nikon D300 and D700 and the Canon 5D Mk II that can definitely write well above 30 MB/s). And there are now many more DSLRs from both brands which use SD cards (the Nikon D90, for instance, can take full advantage of the 133x SD/SDHC cards and can also take at least some advantage of the new 30 MB/s SanDisk SDHC cards, while the more proletariat-priced Canon Rebel XSi can more than take full advantage of an 80x-rated SD/SDHC card). In those newest SLR cameras, it makes no sense to go for anything slower than a 133x-rated card unless the slower cards are substantially cheaper than the 133x cards.

On the other hand, point-and-shoot compact and subcompact digital cameras such as your then-in-possession Canon SD800 cannot take full advantage of anything rated faster than 60x. For those cameras, a modestly-priced SanDisk Ultra II 15 MB/s Edition SD or SDHC card would more than suffice for those cameras.

In between in performance are the point-and-shoot superzooms such as the Canon SX10is, as well as the deluxe compacts such as the Canon G10: Those can take some advantage of the 133x cards, although a true DSLR is faster still. In this case, the aforementioned Ultra II 15 MB/s Edition is ideal for these superzooms and deluxe compacts.

#19

[QUOTE=RJL65;2177539]…And SanDisk’s high-speed cards are among the fastest in their SD speed classes; not surprisingly, the 4GB “Class 4” SDHC Ultra II 15 MB/s Edition (which tested out to about 13MB/sec) outperformed both of the “Class 6” SDHC cards from A-Data and ATP that I had on hand (which tested out to only about 11MB/sec).[/QUOTE]

I am correcting myself on the labeling of the SanDisk Ultra II 15 MB/s Edition cards:

Apparently, some of the newer packages of these cards are now labeled “Class 2” instead of “Class 4.” I just picked up an 8GB version of that card so labeled (Class 2 instead of Class 4).

#20

[QUOTE=RJL65;2178595]I am correcting myself on the labeling of the SanDisk Ultra II 15 MB/s Edition cards:

Apparently, some of the newer packages of these cards are now labeled “Class 2” instead of “Class 4.” I just picked up an 8GB version of that card so labeled (Class 2 instead of Class 4).[/QUOTE]

I have tested the 8GB Ultra II. Although that card has a faster read speed, its write speed is actually noticeably slower than its 4GB sibling - just under 10 MB/s in camera versus about 13 MB/s for the 4GB Ultra II. The newer samples (shipments) of the 4GB Ultra II might also have been similarly castrated. In either case, the results for the 8GB Ultra II makes this line the fastest Class 2 card on the market.