OESF Portables Forum
General Forums => General Discussion => Topic started by: cmisip on November 21, 2004, 09:33:23 am
-
I would like to report my experience with the use of the kingmax 60X SD card which I recently aquired. I created a swapfile on it that is 32 meg in size. Fired up freeswan, shorewall, wireless card and then opera. I went to edealinfo.com and all the images on the site loaded. I was able to scroll up and down the entire page at a fast clip ( some very minor pauses) but opera stayed full on crash free. While not a true benchmark I think this is a pretty good real world test of the speed of the SD card. It is a lot more stable than swapfile on nfs ram. I had not considered SD card speed as a factor in buying my first Sandisk 256 MB card. I would like to advise everyone with a zaurus 5600 that speed of the SD card should be a primary concern as you are going to need a swapfile. One of the reviewers on the newegg site stated that this card is almost as fast as ROM. Don't know if this is true but it sure feels like it is. And I am in no way affiliated with newegg or kingmax.
-
Me too,
just got a Kingmax 1 gig SD card to replace my Kingston 256mb SD card and everything seems to be a lot more zippy. Hard to prove conclusively, but definitely seems faster.
A related question, is their anyway I can create a 64mb swap file as Cacko only gives me a maximum option of 32 mb
I also have not links with Kingmax.
Bob
-
bob-london - search for swap file... you can use the dd command to create as large a swap file as you like and then you mount it.
-
Thanks,
In answer to my own Q this thread currently running elsewhere on this forum explains exactly what I wanted to do.
https://www.oesf.org/forums/inde...?showtopic=8666 (https://www.oesf.org/forums/index.php?showtopic=8666)
Having tried the Kingmax 1gb for a bit with a 64mb swapfile, I reckon I can definitely confirm significant improvements.
This is maybe because I've got a C700, but now I can set ompi, kopi & kapi to fast load, open them all up and then open up opera and everything remains snappy.
Definitely couldn't do this before.
Recomend this as an upgrade at least to those owners of C700s out there.
How much faster than regular SD cards are the Kingmax ones ?
-
How much faster than regular SD cards are the Kingmax ones ?
Well, although the packaging says 60x, I don't really know what "x" is supposed to be.
But the transfer rate of regular SD cards is generally quoted as 2MB/sec and the highspeed SD cards gernally state 10MB/sec.
I cannot remember if this (2 & 10) is megabytes or megabits per second, not do I recall if this is write or read speed (my guess is read speed).
Anyway, this leads me to believe that they average to be 5 times faster. (you can see where I don't get where "60x" comes from)
I wonder what the top-end spped of the Z is for data transfer to these cards? (Meaning at what point does it not matter the speed fo the card as the Z becomes the bottleneck?)
-
x is the transfer rate of a single speed cd-rom drive: 150kbps
With flash memory cards, if there is only one advertised rating it usually refers to write speeds
so a 60x rated sd card should be able to sustain 9mbps writes.
all in bits
-
all in bytes that should be, and sleeping i should be
-
In a previous topic I was wondering about SD speed, and kopsis did a great mini benchmark analysis, and mentioned some useful stats on SD/MMC throughput:
https://www.oesf.org/forums/inde...l=2gb%20sd&st=0 (https://www.oesf.org/forums/index.php?showtopic=7730&hl=2gb%20sd&st=0)
<paraphrashing>It seems with the Zaurus line, even though the cards are rated really high, you don't see performance anywhere near theoretical max, and the performance is highly CPU-bound for SD... </paraphrasing>
Should be enough for video, I decided, but CF seems to be in a higher league.
-Daniel
-
In a previous topic I was wondering about SD speed, and kopsis did a great mini benchmark analysis, and mentioned some useful stats on SD/MMC throughput:
If anyone is interested in running the bonnie++ benchmark that I used against one of these shiny new SD cards, you can grab the binary attached to this post or download it from http://kopsisengineering.com/bonnie.gz (http://kopsisengineering.com/bonnie.gz). I did't bother with an ipk since it's only one file. Just copy bonnie.gz to your Zaurus and from a shell do
gunzip bonnie.gzMake sure that your Zaurus is connected to AC power and that you don't have it set to suspend or turn on screen savers for at least 20 minutes. To kick off the benchmark on an SD card, go to a shell and run the command:
./bonnie -d /mnt/card -f -b -n 0Then go take a nice long coffee break
Compare to my results (https://www.oesf.org/forums/index.php?showtopic=7730&view=findpost&p=47162) and post your findings. I'm very curious to find out if results from some newer SD cards validate my conclusions.
-
i'm running it now against a kingmax 60x 1gb card on a 760 w/cacko 1.21b
-
It's been running for over 40 minutes now. One thing I've noticed is a large number of error messages spewed out from the sd driver:
sd_wait_for_ready: wait card busy time out. status=0700
sd_wait_for_ready error! in sd_write_multiple_block() - 5
pxa_sd_startclock: clock start time out... continue!
I may not get to post my benchmarks run until later.
the sd driver errors seem a bit odd, looking into the system, I see about 55% cpu utilization on kjournald and another 30% on bonnie++
Which file system did you test against?
-
I picked up a Kingmax Platinum 1gb SD card with my 860, and I was suprised at how fast it "felt". I will be very interested in seeing those bonnie results on it.
It is a bit of a strange card - Purple in color and as thin as an MMC card - definitely much thinner than my Lexar SD cards.
-
My first Kingmax 60x 1GB SD card had all kinds of problems (probably caused by something silly I did inadvertently), so I returned it. I just ordered a second one, and should have it in my grubby hands tomorrow for benchmark fun.
-Daniel
-
Which file system did you test against?
My posted data was gathered on a C760 running Cacko 1.21a. The SD card tested was a basic 512MB Lexar formatted ext2. On the same machine I've tested a variety of CF cards with FAT16 and FAT32 formats. I was also able to run this version of Bonnie on a SL-5500 with Sharp 3.x ROM and a 512MB SanDisk CF formatted FAT16.
Do make sure that when you test you use the "-n 0" switch. I had to edit my original post to add that. Without that switch Bonnie does some file creation tests that generate thousands of files. Those tests take forever, aren't terribly useful, and occasionally cause problems on the Zaurus.
-
here's some results:
c760 / cacko 1.21b / kingmax 60x 1gb sd / ext3
$ ./bonnie -d /mnt/card/tmp/ -f -b -n 0
Writing intelligently...done
Rewriting...done
Reading intelligently...done
start 'em...done...done...done...
Version 1.03 ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
zaurus 300M 1447 21 578 3 856 22 29.3 92
zaurus,300M,,,1447,21,578,3,,,856,22,29.3,92,,,,,,,,,,,,,
c760 / cacko 1.21b / kingmax 60x 1gb sd / ext2
$ ./bonnie -d /mnt/card/tmp/ -f -b -n 0
Writing intelligently...done
Rewriting...done
Reading intelligently...done
start 'em...done...done...done...
Version 1.03 ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
zaurus 300M 1530 41 565 31 833 18 26.9 96
zaurus,300M,,,1530,41,565,31,,,833,18,26.9,96,,,,,,,,,,,,,
c760 / cacko 1.21b / kingmax 60x 1gb sd / ext3 again without "-n 0"
$ ./bonnie -d /mnt/card/tmp/ -f -b
Writing intelligently...done
Rewriting...done
Reading intelligently...done
start 'em...done...done...done...
Create files in sequential order...done.
Stat files in sequential order...done.
Delete files in sequential order...done.
Create files in random order...done.
Stat files in random order...done.
Delete files in random order...done.
Version 1.03 ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
zaurus 300M 1324 10 553 3 796 19 26.4 86
------Sequential Create------ --------Random Create--------
-Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
16 29 24 7786 76 48 2 31 25 666 96 26 6
zaurus,300M,,,1324,10,553,3,,,796,19,26.4,86,16,29,24,7786,76,48,2,31,25,666,96,
6,6
-
uhm...
anyone can translate the above report into something for non-sd-and-fs-experts ? Is then the kingmax x60 1gb sd worth ?
thanks,
-
anyone can translate the above report into something for non-sd-and-fs-experts ? Is then the kingmax x60 1gb sd worth ?
For the 1GB Kingmax 60x SD card:- block write speed: 1530 KB/sec at 41% CPU load
- block read speed: 833 KB/sec at 19% CPU load
This is nowhere near the card's 60x theoretical max of 9000 KB/sec and supports my conclusion that SD card performance on the Zaurus is bound not by the performance of the SD card but by the SD interface on the Zaurus itself.
Results for a run-of-the-mill PNY 512 MB SD card (on an almost identical Zaurus):- block write speed: 787 KB/sec at 22% CPU load
- block read speed: 1341 KB/sec at 66% CPU load
Those results seem to indicate that the Kingmax may actually sacrifice read performance for improved write performance. That's great for cameras that do mostly write operations ... not so great for a computer filesystem that sees mostly read operations. It also validates the post that started this thread which was that using a swapfile on the Kingmax seemed to give good performance. A swapfile typically sees an equal number of reads and writes. The average throughput for the Kingmax is 1182 KB/sec and 30% load. For my PNY its 1064 KB/sec and 44% load. Clearly the Kingmax will perform better as swap space.
Bottom line (to me) is that if you're buying an SD card exclusively for the Zaurus, don't pay a premium for "high-speed" SD cards unless you have an "atypical" usage pattern (swap file or writing lots of data). In general, you won't see much benefit over their "normal-speed" siblings and you may actually do worse where it typically counts (read performance). If you're going to use the card in multiple devices (Zaurus & PC, Zaurus & camera) then the high speed might be worth a few extra $$.
-
The Kingmax is one of the cheapest 1GB cards, so no extra $$ is involved in this case. I just payed $70 (plus s/h,taxes) for one. Too bad about the read speed though.
-Daniel
-
Here's mine does anyone understand it ??
$ ./bonnie -d /mnt/card/Tunes -f -b -n 0
Writing intelligently...done
Rewriting...done
Reading intelligently...done
start 'em...done...done...done...
Version 1.03 ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec
%CP /sec %CP
zaurus 300M 1173 7 533 3 780
17 30.6 91
zaurus,300M,,,1173,7,533,3,,,780,17,30.6,91,,,,,,,,,,,,,
$
-
I think the C700 has much lower "ram" than the 760 & 860 so my experience of using the Kingmax with a 64mb swap probably gets a double benefit i.e. double the write speed, double the swap ( I also only previously had a 32mb swap)
-
Here's mine does anyone understand it ??
$ ./bonnie -d /mnt/card/Tunes -f -b -n 0
Writing intelligently...done
Rewriting...done
Reading intelligently...done
start 'em...done...done...done...
Version 1.03 ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec
%CP /sec %CP
zaurus 300M 1173 7 533 3 780
17 30.6 91
zaurus,300M,,,1173,7,533,3,,,780,17,30.6,91,,,,,,,,,,,,,
$
The trick to reading the bonnie++ output is to ignore everything except the last line of comma seperated fields. The magic decoder ring for that last line is:- Field 1 = host name (zaurus)
- Field 5 = block write KB/sec
- Field 6 = block write CPU %
- Field 11 = block read KB/sec
- Field 12 = block read CPU %
So your results are block writes at 1173 KB/sec - 7% CPU load, block reads 780 KB/sec - 17% CPU load.
-
Here's my results from a panasonic 1GB ext2:
$ ./bonnie -d /mnt/card/ -f -b -n 0
Writing intelligently...done
Rewriting...done
Reading intelligently...done
start 'em...done...done...done...
Version 1.03 ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
zaurus 300M 1290 6 504 61 749 57 20.4 95
zaurus,300M,,,1290,6,504,61,,,749,57,20.4,95,,,,,,,,,,,,,
So I guess that makes it 1290kbs@6% CPU write, 749kbs@57% CPU read?
Reads seems too high in CPU, I'll probably have to rerun it again, my screen saver did pop up a few times (didn't realise the benchmark would take SO long to run!).
[edit]
Here's the rerun:
$ ./bonnie -d /mnt/card/ -f -b -n 0
Writing intelligently...done
Rewriting...done
Reading intelligently...done
start 'em...done...done...done...
Version 1.03 ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
zaurus 300M 1338 70 505 9 750 18 20.4 95
zaurus,300M,,,1338,70,505,9,,,750,18,20.4,95,,,,,,,,,,,,,
Now it says writes take 70% CPU time?!? yet another run time...
Version 1.03 ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
zaurus 300M 1318 24 505 9 759 18 20.6 96
zaurus,300M,,,1318,24,505,9,,,759,18,20.6,96,,,,,,,,,,,,,
ok this seems better?!?
[/edit]
-
Just got a Transcend 1Gb 60x SD and now feeling rather foolish after reading that an Ultra-type card isn't going to give me much of an I/O improvement.
Anyway, here's the results from bonnie :
zaurus, 300M,,, 1378, 23, 838, 10,,, 1531, 24, 43.8, 92,,,,,,,,,,,,,
That's out of the box, fat formatted. I can't vouch for its reliability or how fast it actually feels yet, because I've only just got it.