Xbit labs pushes the envelope once again! It's nice to see such a practical review, far outside the mainstream of most hardware enthusiast sites, and much, much more detailed than consumer computing magazines.

My card reader of choice is a Mitsumi FA-404M, a 3.5" internal with integrated floppy drive. The chip appears to be a GL819 with firmware 93.25 -- so this gives a good picture of its capabilities already, though I can't help but wonder how a v93.25 would perform with MMCplus cards; too bad I don't have any around.

Great write up but why no mention of DMA? The newer, faster (266x, 300x, etc.) compact flash cards acheive those results by using Ultra DMA interfaces. The older cards and readers are limited by the PIO interface they use.

The 6377 based readers obviously use UDMA. I have a Sandisk extreme that the vender claimed was UDMA but my result matched these; PIO speeds only.

As far as I know, no SD or SDHC cards or readers go above PIO speeds.

Because it is not very popular, I have never looked into MMC. Some of those results were in the UDMA speed territory. Does the MMCPlus standard include support for UDMA?

So errrrr....which internal card reader do i buy ? :|

Interesting article BUT why do an article about a product(s)that when you want to buy it it's no where to be found on the net.
Case in point the Microsonic CR 90.

my vaio lappy enables windows vista ready boost with a memory stck pro duo 2gb from sandisk and i use da inbuilt card reader....i dont hav a problem with it

Check your SanDisk readers... There are firmware updates that add SDHC to some of them, including most versions of the one listed here as "SanDisk USB." (The correct name for that product is "SanDisk ImageMate® 5-In-1 Reader/Writer.") Two versions of the 12 in 1 reader also gain SDHC support with a firmware update.

Please update/correct this article with the model of the Sandisk readers tested and its frimware. It matters. There are presently 5 variants of both ImageMate products that could be floating around the supply chain.

The same may be true of some others that didn't support SDHC in these tests. A firmware update may enable SDHC and/or affect the transfer rates. (As noted in one reader when firmware verisons were tested and the older frimware was faster.)

Another 10/10 article by Xbit-Labs.
This is how a testing should be done: once again Xbit-Labs is defining the rules, not following them.

My Kingston FRC-HS219/1 has a GL826 + PMC 0732, while my Kingston FCR-HS215/1 has a GL819 + PMC 0628.

Differences between the two are huge: HS215 has tons of SMD resistors and capacitors, high quality soldering point, nice PCB. HS219 seem to be a low (very low) chinese quality item, with dirty soldering and bad quality overall.

This also explains why sometimes I have problems with HS219, while HS215 always worked without problems.

Funniest thing is, while PCB-quality of HS219 is lower, external plastic is prettier!

You might like to double check the review.. Seems to be some confusion between the AU6362 and a chip called the AU6263.

Besides that a very intersting read, I'm using a Transcend device (not sure of the exat model number) that luckily has the AU6362 with rev 1.00 firmware ;)

was alright, 7/10.
should've included the Scythe's YD-8V08 though, solid 16 in 1 card reader.

I can see lots of recommendations for this test on the dpreview storage forum.

How can you test 27 outdated readers? There is not a single modern controller model used.

Where in Europe can i buy the Microsonic CR09 card reader?