Drobe :: The archives
About Drobe | Contact | RSS | Twitter | Tech docs | Downloads | BBC Micro

UniPod speed tested

By Andrew Hill and Chris Williams. Published: 2nd Jun 2004, 21:08:53 | Permalink | Printable

IDE and ethernet with a need for speed

Small unipod photoAnnounced at the very end of March, the UniPod is the new combo podule from STD that features two USB ports, an IDE interface and a 100MBit ethernet network port. Seeing as the UniPod was touted as "the only RiscPC podule you'll ever need", we've spent the past week or so speed testing the upgrade to see how it fares.

Now, this article isn't intended to be an in-depth study, nor is it a comprehensive review of the product - it merely looks at how fast the UniPod's IDE and network interface can transfer data, compared to existing hardware. The data collected and presented below will vary depending on individual setups, but the figures should give an overall indication of performance. As always, we've tried our best to be as fair and accurate as possible, but if you spot any anomalies, feel free to point them out.

The testing begins
A RISC OS 4.39, StrongARM powered Kinetic RiscPC, with a Radeon Viewfinder and 192MB of RAM fitted, was used to test the UniPod, while Dave Holden's HDspeed application was used to test the speed of the podule - a copy of which can be found on the VirtualRiscPC CD. HDspeed employs a number of techniques to test a hard disc and the interface between the hard disc and the host computer, and we've presented the most important results below. As we kept the hard disc under test and the host computer constant, the performance of the interface will be apparent. Incidentally, the hard disc used was a 60GB Hitachi Deskstar 160GXP.

Firstly, a few notes on what each test does and what the outcome represents:

Sequential byte access
There are various approaches used by software when saving and loading your information to and from storage media. The information can be saved as a stream of write operations, by saving one byte (or character) after another in a sequential manner. It's not terribly efficient, but it can be used when saving information that is being constantly updated or is generated 'on the fly'. The sequential byte access test investigates the interface's performance in this area and reports how much time each process took in centiseconds - where one centisecond is a hundredth of a second.
Block access
Another method of saving and loading information is to transfer the data in a block. This is more efficient, but requires the application to have all the information together beforehand. This is the preferred method when saving and loading large files like images and documents, so it gives a good indication of loading and saving speeds in the real world. The block access test investigates the interface's performance in this area and reports the transfer speed in kilobytes per second.


The HDspeed testing was first undertaken on the RiscPC with the hard disc attached to the UniPod IDE interface. Then for reference, HDspeed was run again on the RiscPC with the hard disc connected to the internal Acorn IDE interface and again with the hard disc connected to an APDL Blitz IDE card. Next, a Dell Optiplex GX260 1.8GHz PC, with a 100/1000MBit network card, was connected via a cross-over cable to the RiscPC's 100MBit network interface (NIC) on the UniPod. HDspeed then tested transfers to and from a LANMan mount to briefly investigate real world network transfer speeds.

Finally, before we even begin, one of the most limiting factors is going to be the RiscPC's legacy podule bus, which by design can achieve a maximum data throughput of approximately 6100KB per second.

The results

UniPod IDEInternal IDEBlitz IDEUniPod NIC
Sequential byte access
(centiseconds)
Read 50KB12.84.312.719.4
Write 50KB23.639.223.336.0
Read 500KB130156130203
Write 500KB241400238368
Block access
(KBytes/second)
Save 50KB241511932487925
Load 50KB3496100635711136
Save 2MB3543168437371381
Load 2MB3961189642481754


Essentially, the UniPod IDE interface sustains around 3.5MB/s for saving files in blocks, and 3.9MB/s for loading files in blocks. The network interface was shown to achieve 1.3MB/s when saving a file via LanMan and 1.7MB/s when loading a file via LanMan. Other UniPod users have confirmed that they also get around 3.5MB/s when load and saving files to disc, although STD told us that "the environment in which a card runs is obviously very significant for speed tests".

Drobe hardware motifFeedback
During the course of our testing, our results were initially much lower than anticipated as STD expected us to get at least 4MB/s when reading and writing to disc. As a result of our findings and the following investigation, STD discovered a 'safety mechanism' in the UniPod system that forces attached hard discs to run in a low PIO mode, which limits the data transfer speed - we're told that the UniPod documentation is in the process of being updated on how to ensure that your hard disc transfers data at an optimum speed. Plus, a newer version of the not entirely bug free UniPod IDE hard disc formatter and configuration tool, IDETool, and new USB firmware should also be available soon.

Other notes
The UniPod card tested, which incidentally is a developer model, also includes an internal USB header which allows you to use internal USB devices, or to connect USB ports to other parts of the computer; although the device supports a maximum of two USB devices. The IDE connector is also keyed, which means that 80-pin cables that have solid pins to ensure correct keying, can be used without resorting to manual drilling.

Pushing to one side the arguments against why people should indeed bother to upgrade legacy computers like ten year old RiscPCs, the UniPod does offer IDE access that outperforms the internal IDE interface, whilst providing two other important interfaces: USB and fast ethernet.

Update at 15:56 3/6/2004
It should, of course, be noted that the RiscPC's Kinetic card disables DMA operation for the podule cards shown above. Therefore, if you're a Kinetic RiscPC user, the results above should be what you'll get. If you're a non-Kinetic RiscPC user, then you'll see much faster speeds, in the order of 25 to 50 percent faster. When Andrew gets his StrongARM card, we'll post DMA enabled results.

Links

Unipod website

Previous: Iyonix mobo spotted in the wild
Next: RISC OS Adjust Review

Discussion

Viewing threaded comments | View comments unthreaded, listed by date | Skip to the end

What happens if you copy from the network to harddisc (on the podule) or vice versa? Does it go much slower, twice passing the podulebus, or doesn't it need to pass this bottleneck and it therefore faster? Or doesn't that really matter?

 is a RISC OS Userjjvdgeer on 3/6/04 7:09AM
[ Reply | Permalink | Report ]

I don't know who did the tests but it seems from the results that the Blitz was tested without DMA.

With the drive described and DMA enabled block transfer speeds should be at least 50% higher.

 is a RISC OS Userapdl on 3/6/04 7:40AM
[ Reply | Permalink | Report ]

Yes, they probably should have mentioned that in the report, though they did say the computer was a "A RISC OS 4.39, StrongARM powered Kinetic RiscPC" The key word being 'Kinetic'...

 is a RISC OS UserRimmer on 3/6/04 8:30AM
[ Reply | Permalink | Report ]

jjvdgeer: I've not tested it, but everything will have to go across the podule bus twice. All the network interface sees are just packets - doesn't know who they're for, what order they're in and can only buffer a few. All the working out where each packet goes is for the host to do.

 is a RISC OS Usercaliston2 on 3/6/04 9:16AM
[ Reply | Permalink | Report ]

"The Unipod does offer IDE access that outperforms the internal IDE interface"... :cooldude: I'm told that the PC hard disks tend to wear out faster because it's system works the HD harder to retrieve both files/programmes and system instruction all from the HD. :rolleyes: As I understand, RISC OS is more 'Efficient' with it's instruction in Flash Rom, hence the HD doesn't need to work harder giving RISC OS HDs (generally speaking) a longer life span. :-o Therefore, in the article statement I mentioned first, would a high performance IDE in this Unipod work the RISC OS HD faster, thus giving it a shorter life span? :blush: Finally, does a much larger memory (like the 192Mb RAM, 60GB HD in the article) installed on a computer, increase the :sparky: speed performance of the machine? Forgive my comments for I am not technically computer minded, :bowdown: just putting together info as I try to figure all this out! :wink: Cheers, Steve.

 is a RISC OS UserSawadee on 3/6/04 9:17AM
[ Reply | Permalink | Report ]

David - I agree with your point; DMA is disabled with the Kinetic upgrade and the cards may perform differently on a DMA-enabled machine. I can see a case for repeating the tests on a DMA-enabled machine. I can aim to repeat this test when I've got the card and post the tests with DMA enabled; unfortunately my SA card is back in Notts and will take me a fortnight to do the test (unless someone wishes to post me instructions on how to temporarily turn a Kinetic into a normal SA card).

Steve. Hard drive stress is a very difficult subject. It's not the throughput of data which would shorten drives significantly; drive temperature and number of power cycles are the major factors. Having said that, DO NOT use drives for much beyond their warranty period - the consequences of having a dead drive are catastrophic.

Hope all that helps,

Drew

 is a RISC OS Usermd0u80c9 on 3/6/04 11:06AM
[ Reply | Permalink | Report ]

Might also be worth running the tests with an Arm710 card in the machine. I find that I can get much higher block transfers through the Blitz interface (in a DMA slot) than I can with a StrongArm card.

Cheers!

 is a RISC OS UserThe Doctor on 3/6/04 3:14PM
[ Reply | Permalink | Report ]

Whilst I agree that you probably would see a difference, my whole purpose was to demonstrate the effects on current hardware, and I wouldn't consider an Arm 700 RPC as 'current'. Also whilst this is attempting to look at 'real-world' values, it's not a 100% accurate performance indicator for all machines in all circumstances.

Whilst I think the Unipod is excellent value for money, a second-hand SA card would add more to your machine for that money, so my advice to ARM 700 RPC users is simple: use the money on a SA first, and if you get the money again, come back for the Unipod if it suits your needs. If they desperately need, eg, USB or an NIC, then the choice of a Unipod would be obvious as it definitely offers the best value-for-money, but in those cases speed comparisons would be a moot point.

 is a RISC OS Usermd0u80c9 on 3/6/04 5:30PM
[ Reply | Permalink | Report ]

My experience with UniPod shows the IDE is slower with DMA on and the ethernet only gets about half the rate quoted above as it keeps stalling. Clearly YMMV. STD are aware of the issues, however. Here's hoping for the new firmware soon! :-) Even ignoring the ethernet, the USB and IDE on one podule provide a good space saving and nice new features. If STD ever release the developer docs for the addon features and parallel port we may also see it really become _the_ podule to have (after ViewFinder :-) )

 is a RISC OS Userimj on 3/6/04 7:06PM
[ Reply | Permalink | Report ]

I was looking at the picture (no plate? how on earth do you screw this podule in? ;-) ) and saw the internal USB header.

Read the text.

Are you limited to 2 USB ports internal *and* external?

Does !MassFS now support USB mass storage devices properly?

i.e. could you fit a CDRW + card reader or a floppy + card reader to a Risc PC now?

I am curious :-)

 is a RISC OS Userepistaxsis@work on 3/6/04 8:13PM
[ Reply | Permalink | Report ]

There is a plate - it's removed upon shipping, and you screw it on when you take it out the box.

Epistaxis - you are limited to two interfaces TOTAL, ie. two external, or one internal and one external. I have not played with this, and you will have to wait for STD to provide information re this.

IMJ - I didn't find any experience with it stalling, despite regular use, so hopefully this is a firmware issue.

 is a RISC OS Usermd0u80c9 on 3/6/04 8:23PM
[ Reply | Permalink | Report ]

Epistaxis: Yep, MassFS works very nicely. Connect the device and up pop drive icons on the iconbar. Super. Been using it with a flash drive and smartmedia reader from my camera.

md0u80c9: I've had beta firmware since Wakefield, no improvement :unhappy: :crybaby:, but we live in hope. ;-)

 is a RISC OS Userimj on 3/6/04 10:40PM
[ Reply | Permalink | Report ]

md0u80c9:

thanks for the clarification

 is a RISC OS Userepistaxsis@work on 3/6/04 10:44PM
[ Reply | Permalink | Report ]

Drew. It was pointed out to me that PC Hard Disks (tend) to wear faster, I think they mean one factor is that the PC HD is used by the machine a more than a RISC OS machine does. (I personally wouldn't know - but others say to me these things). :unhappy: Other major factors as you pointed out - such as temperature, an obvious point I was not aware of as to the importance of cooler running RISC OS machine (not forgetting QUIETER)! Also from what you are suggesting, it really pays to renew your Hard Drive before it ages or is due to 'run to the ground' (die!). Thanks! :cooldude: Steve

 is a RISC OS UserSawadee on 4/6/04 8:28AM
[ Reply | Permalink | Report ]

Whilst it's logical to suggest the more you use it, the quicker it wears out, AIUI this isn't the case with normal running of desktop drives.

Power cycling is the major issue; the more you power cycle the drive, the quicker it dies. From the drive, I fished out the number of power cycles my drive has done in just a few months - it was astonishing.

FWIW, I also learnt that in my RISC OS box on the top slice, my drive runs at a toasty 47 degrees. I actually cut it down (from 51 degrees) with Unipod. That's not due to anything clever - merely the fact that with it I got a rounded IDE cable in, and removed a USB card, NIC and IDE card and replaced it with a single board, so more air throughput was possible. Those figures are presumed to be correct, but the software used is still in development.

However, as I've said, the take-home message is NEVER use old drives for anything other than non-essential storage (NOT BACKUPS, because when you come to reach for your backup...) or doorstops - otherwise you ask for what you get.

 is a RISC OS Usermd0u80c9 on 4/6/04 11:39AM
[ Reply | Permalink | Report ]

I have got my backups networked on two other RISC OS computers. But I'm noting the importance of keeping these machine cool. With 11 RISC OS machines in two adequately sized small rooms and up to 20 children working on the at times, the doors and the windows are open to vetilate air circulation and this cooling may help the Hard Disk life? Not to forget, I must keep the school's water heater tap valves down to a minimum during winter as well. :blush: The (PC)computer room in the next building however, gets so hot in the warmer summer days I sweat easily (many others complain about the computer sauna room). Almost 3 years old, a number of the new PC's have had HD and other major problems already. :-o Our school needs airconditioning for the main computer room, what's an ideal temperature for about 30 computers in an old design classroom just big enough for them? :cooldude: Cheers, Steve.

 is a RISC OS UserSawadee on 4/6/04 12:17PM
[ Reply | Permalink | Report ]

Sawadee: Can a RISC OS machine with a UNIPod boot from a USB-stick? Yes, than you can stop using Hard Disks.

> what's an ideal temperature for about 30 computers in an old design classroom just big enough for them? Too cold for the students : 10-15 'C (the lower the beter)

 is a RISC OS Useregel on 4/6/04 1:54PM
[ Reply | Permalink | Report ]

Like beer, they work better cold.

But obviously not sub-zero, as like beer, they can freeze :o).

 is a RISC OS Usermd0u80c9 on 4/6/04 8:58PM
[ Reply | Permalink | Report ]

BEER!! :-o Now we're talking English (not tech jargon). I'm an Aussie, chilled beer's the one and I clearly understand the point about sub-zero and computers now. ;-) But not a problem understanding those tips in terms of beer, we Aussies chill and care for our beer with utmost pride and care. :acorn: Like one of the Acorn !Sibelius7 "Tips of the Day", it says 'don't water your computer'. So I don't think I'll chill my computer (like beer) on ICE!! :cooldude:

'Reply to egel:' Obviously as you suggest, the cooler the better - with ideal being maybe 10 - 15C ? In the summer months here in New Zealand, a computer room that has virtually no cooling apart from a couple of ceiling fans and not to forget the massive body heat created here, your ideally suggested temperature is sadly about half what we experience at times!! I can see our ICT computer teacher submitting a :bowdown: BIG ask for airconditioning after the huge millions spent on our school the past four years. But then again I would image the airconditioning may not outweigh the costs over years of repairs to overheated computers (and overheated children and teachers!). ;-) Some of the points in the article were interesting about allowing (if possible) airflow around the components in the computer. Now would this argue a point against the thought of buying a compact slimline computer case?? :rolleyes: Cheers, Steve

 is a RISC OS UserSawadee on 5/6/04 3:48AM
[ Reply | Permalink | Report ]

sawadee:

If you're worried about heat you could get some teeny weeny fans mounted in the ventilation slots in front of the Hard Disk - set to blow over the drive.

The RiscPC never did pressurise too well - especially once you went beyond a couple of slices...

 is a RISC OS Userepistaxsis@work on 5/6/04 2:41PM
[ Reply | Permalink | Report ]

Re: Air Condtioning Sometimes, "rapid air extraction" is all that's needed, (i.e. rather than "air conditioning", which may be cheeper and more environmentally friendly. i.e. Simply suck the hot air out rather than try to cool it down.

 is a RISC OS Usermartin on 6/6/04 12:04AM
[ Reply | Permalink | Report ]

 is a RISC OS Userjlavallin on 6/6/04 1:46AM
[ Reply | Permalink | Report ]

If an airflow is sufficient, then I have just about that in the summer months when I open windows and doors, there is a decent draft flow through my clasroom area. But as for the computer room, maybe a few of it's useless large lower windows replaced with decent ones that open enough for a good draft with extra fans??? Airflow in a room is often sufficient when you think about it. ;-) Thanks, Steve.

 is a RISC OS UserSawadee on 6/6/04 7:28AM
[ Reply | Permalink | Report ]

Re RiscPC cooling [link] I've done this mod on all of the RiscPC's I have owned.

On a two slice machine with only podules in the second slice and a PC card on the motherboard, I blanked off the cooling slots on the top slice thus forcing air out only via the slots in the bottom slice. The machine became more stable in hot weather as a result of these changes.

My A3010 used to crash a lot in summer when the ambient temp went above 25'C ;o)

 is a RISC OS Userblahsnr on 7/6/04 10:37AM
[ Reply | Permalink | Report ]

Could you provide some pictures on how to do that blahsnr? In fact, thinking about it - Ian's just posted the combo-drive how-to. Why not spruce up your page with some piccies, and it could go onto the Drobe hardware archive???

 is a RISC OS Usermd0u80c9 on 15/6/04 2:55PM
[ Reply | Permalink | Report ]

Has Andrew got his RPC back yet ? I'd love to see the DMA-enabled results .. especially now STD are selling them again !

 is a RISC OS UserJohnB on 15/7/04 8:57AM
[ Reply | Permalink | Report ]

Not the RPC that was the problem - I needed the SA card. Now I've got that and...the drive dies. Under warranty, so I'm waiting for it to get back. Will post - I often need prompting though as I'm a little snowed at the moment. BTW - any ideas how to turn a Kinetic into a SA?

 is a RISC OS Usermd0u80c9 on 15/7/04 5:45PM
[ Reply | Permalink | Report ]

A jumper setting on the kinetic will make it a normal SA card.

 is a RISC OS Usermicken on 17/8/04 3:53PM
[ Reply | Permalink | Report ]

Please login before posting a comment. Use the form on the right to do so or create a free account.

Search the archives

Today's featured article

  • Pace and Sega announce Dreamcast joint venture

     Discuss this. Published: 29 Jan 2001

  • Random article

  • PNG developers kit from RISCOS Ltd.
    Released exclusively to Select users
     12 comments, latest by moss on 3/10/02 5:22PM. Published: 2 Oct 2002

  • Useful links

    News and media:
    IconbarMyRISCOSArcSiteRISCOScodeANSC.S.A.AnnounceArchiveQercusRiscWorldDrag'n'DropGAG-News

    Top developers:
    RISCOS LtdRISC OS OpenMW SoftwareR-CompAdvantage SixVirtualAcorn

    Dealers:
    CJE MicrosAPDLCastlea4X-AmpleLiquid SiliconWebmonster

    Usergroups:
    WROCCRONENKACCIRUGSASAUGROUGOLRONWUGMUGWAUGGAGRISCOS.be

    Useful:
    RISCOS.org.ukRISCOS.orgRISCOS.infoFilebaseChris Why's Acorn/RISC OS collectionNetSurf

    Non-RISC OS:
    The RegisterThe InquirerApple InsiderBBC NewsSky NewsGoogle Newsxkcddiodesign


    © 1999-2009 The Drobe Team. Some rights reserved, click here for more information
    Powered by MiniDrobeCMS, based on J4U | Statistics
    "Don't suppose you could remove that article, just for a few hours?"
    Page generated in 0.0942 seconds.