I use a Mac and not a PC


No idea why they get such low results but then again there was something linked above which gave very similar results, for me it is the same if I just 'time dd' very consistent...
Yes, but the data HAS to be uncompressable. Does the aja system test actually come with a 4G playable video file which is used for the test or does it write the equivalent of a 4G file? The data has to be entropy encoded otherwise it's not uncompressable. If this doesn't hold, the speed will be much higher as in the case of a dd....
I doubt it's a PC/Mac thing. Fundamentally the performance is limited because to achieve the high speeds, SF controllers compress the data as it is written to the flash. This is all done in the drive firmware and if you can't compress the data, it will go slower. This is what all benchmarks of the SF controllers show.
I can google any number of tests on the Vertex 3 and they all show the same thing: 500MB/s read, ~300MB/s write for uncompressable data. A random one here:
http://benchmarkreviews.com/index.p...sk=view&id=717&Itemid=60&limit=1&limitstart=5 or this one:
http://www.anandtech.com/show/5785/corsair-performance-series-pro-256gb-review/4 (you'll see it half way down the second performance table).
The Vertex 3 is an attractive proposition at £185 whilst the Perf. Pro is £270, but to claim it is faster when dealing with compressed data is not the case.
FWIW, I've just run the AS-SSD benchmark on my 90G Vertex 2 in my i7-2600K and have got a paltry 146M read, 65M write. It does have a nice .2ms access time though.
I'll rerun the test on the Perf. Pro we have in the laptop when Ms arad85 is finished with it.