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

Profile for stdevel

ContactAbout me
Email:
stuart@stdevel.co.uk
ICQ:
AOL:
Yahoo:
Username: stdevel
Realname: Stuart Tyrrell
About me:
Homepage: http://www.stdevel.co.uk
Face/Logo:
Comments posted:67 (show all)

Recent comments

On RISCOS.org joins pixel ad bandwagon:

flibble:

Contract and Cheque in post, artwork with the artist now B-)

 is a RISC OS Userstdevel on 19/1/06 8:28PM
[ Reply | Permalink | Report ]

On VirtualRiscPC network upgrade pulled:

If anyone does have access to or thoughts on a decent set of *user* benchmarks (I will be publishing technical benchmarks, but they take a /lot/ of interpretation - if one series of low-level operations are half the speed of another machine, and another series twice as fast, what's the comparative speeds of the machines?), I'll be happy to run them. The marks that Steffen Huber published the results between the Iyonix and Omega are great (well, OK, maybe not so great for one machine) - but at the moment he's not able to publish the source files as originally promised. We (RISC OS users) need a set of suitable, reasonably independent set of user-benchmark files which can be distributed and peer-reviewed, and I think given comments above I'm not the person to arrange this.

The "results" I posted regarding a simple "user" test above are stopwatched (or at least at the moment that's all I'm saying they are), however I will always err on the side of caution. The reason why I'm claiming they are stopwatched at this stage is it really isn't important whether one machine is faster than the other to within 5%, the pertinant facts are some tests are similar, others faster, others slower and this can be used as an illustration of the type of actions which are likely to give certain results in other tests. If I wanted to claim actual speeds of machines I'd choose or make up some suitable benchmarks which favoured the A6, rather than choosing the only publically available user-benchmark that we seem to have.

However, again, when someone asks directly "Will VARPC ever match the speed of the Iyonix", am I to sit back and watch conjecture? The A6 and its ilk will live or die on their merits alone - that's the reason why in the first days I didn't even mention the processor speed of the machine - until the uninformed started suggesting I had something to hide. I'm sure it won't be long before someone runs Paul Vigay's test on some random XP2500+ shared-memory over PC133 box and claims that I must be massaging results......

 is a RISC OS Userstdevel on 02/11/03 5:08PM
[ Reply | Permalink | Report ]

On VirtualRiscPC network upgrade pulled:

Ah, OK. I see.

So David asks "are we likely to see VRPC run faster that the Iyonix", and I'm supposed to sit back and watch the conjecture and random backbiting rather than answer the question with information which is pertinant, because a statement of fact might be interpreted as advertising.

Sorry, I hadn't realised that was the objection - I'm primarily an engineer and I expect information to be digested and stand in its own right rather than massaged into a "message". If my information is going to be interpreted as advertising I might as well only post opinions which present the A6 in a "perfect" light rather than also showing its weaker points, surely?

I'm not into the whole marketing-speak like "The fastest ABC-OS machine is a great development platform. We're pleased to see XYZ on ABCMachine". I hope vague meaningless comments we've seen like this recently aren't going to mean that every scrap of data from all of us is going to be viewed with equal suspicion.

 is a RISC OS Userstdevel on 01/11/03 6:34PM
[ Reply | Permalink | Report ]

On VirtualRiscPC network upgrade pulled:

Mark - I'm sure that you'd like to take the opportunity to clarify the meaning of your last post. In the meantime you'll appreciate that it would be suicide for us to make claims in a commercial environment which will be subject to peer review, advertising standards, press comment etc which were anything but presentable and proveable.

I have been extremely careful with statements I have made. If you don't like them, then fair enough, but please take care not to imply that I have not taken due diligence.

 is a RISC OS Userstdevel on 01/11/03 2:36PM
[ Reply | Permalink | Report ]

On VirtualRiscPC network upgrade pulled:

Cor, that's a leading question! I've been looking for sensible "real world" benchmarks. In the absence of Steffen's (he's got copyright and time issues), I had a quick play with Paul Vigay's graphics test on the A6 (currently stopwatch times, I'll release "officially" when I can prove them absolutely). Iyonix times from Paul's website.

Loading image was 2.6 seconds (Iyonix 2.51) Rotating image was 4.5 seconds (Iyonix 9.09) Unsharpen mask was 117 seconds (Iyonix 74.91) Scale was 2.2 seconds (Iyonix 2.26)

The A6 seems to perform faster than the Iyonix when it's "shifting data" with little transforms (eg when you're scrolling around a large DTP document); The Iyonix scores better on cache-based repeated operations where more instructions act on each bit of data (which hits the long pipeline on the A6 reasonably hard).

Your percieved speed of a machine will depend very much on the profile of what you do with it. The A6 subjectively feels slickest for many "user" tasks, but YMMV and I'm biased ;-)

 is a RISC OS Userstdevel on 30/10/03 10:37PM
[ Reply | Permalink | Report ]

Search the archives

Today's featured article

  • David Ruck comments on RM's decision to pull out of education in Northen Ireland

     1 comment, latest by AndrewDuffell on 5/12/03 7:44PM. Published: 13 Jan 2001

  • Random article

  • Get your files mobile with PsiFS 1.4

     Discuss this. Published: 6 Mar 2001

  • 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
    "RISC OS needs a frequently updated news site that has all the news, not just a carefully selected sample. Currently this simply isn't Drobe"
    Page generated in 0.1007 seconds.