log in | register | forums
Show:
Go:
Forums
Username:

Password:

User accounts
Register new account
Forgot password
Forum stats
List of members
Search the forums

Advanced search
Recent discussions
- Latest hardware upgrade from RISCOSbits (News:)
- WROCC November 2024 talk o...ay - Andrew Rawnsley (ROD) (News:2)
- October 2024 News Summary (News:3)
- RISC OS London Show Report 2024 (News:1)
- RISC OS London Show 2024 - pictures (News:2)
- RISC OS London Show 2024 - Notes from the talks (News:)
- RPCEmu 0.9.5 (Gen:2)
- Late breaking news from RISCOSbits (News:)
- ROD updates RISC OS Direct release (News:)
- What will R-Comp be showing N.Ex.T? (News:)
Related articles
- Omega LegPuller at ROUGOL meeting
- Iconbar in update shocker!
- Wakefield 2003 - the preview
- ArtWorks 2 uses Iyonix advantage
- Aemulor Released
- The Iyonix has been launched!
- RISC OS 2002 show
- Wakefield 2006 show report
- Wakey Wakey, it's show time again!
- 50,000 shares, Iyonix Select and a Belated Happy Birthday
Latest postings RSS Feeds
RSS 2.0 | 1.0 | 0.9
Atom 0.3
Misc RDF | CDF
 
View on Mastodon
@www.iconbar.com@rss-parrot.net
Site Search
 
Article archives
The Icon Bar: News and features: Omega and Iyonix compared
 

Omega and Iyonix compared

Posted by Phil Mellor on 01:47, 19/10/2003 | , , , , , ,
 
Steffen Huber (author of CDBurn) has published a comparative review of the Iyonix and Omega on his web site - and includes a very interesting table of benchmark results. The main articles are currently in German, so you might find AltaVista's translation service useful.

Rather than simply performing number crunching, the benchmarks test for stuff users are likely to do, such as redrawing the ArtWorks apple, compressing files with SparkFS, and rendering text in Insignia. In most tests the Iyonix clocks in twice as fast as the Omega (perhaps unsurprising considering the 600Mhz vs 300Mhz processors) and the Omega runs a tad faster than a StrongARM Risc PC.

Worringly, converting a large JPEG to PNG in Creator is far slower on the Omega than a Risc PC - although the Iyonix isn't that much faster than the Risc PC either. The Omega also performs badly in a file compression test - which may indicate issues with the UDMA hard disc interface need to be resolved. For balance, apparently the Omega performs better than the Iyonix in certain situations, according to RISC OS Ltd (see Drobe's show report).

Of course, the Omega's design allows the potential for future enhancements - along with such trivial matters like networking, USB and XScale support. At least the floppy drive jolly well works now.
 

  Omega and Iyonix compared
  jmb (01:49 19/10/2003)
  Steffen Huber (20:38 19/10/2003)
    richcheng (15:37 20/10/2003)
      richcheng (15:38 20/10/2003)
        monkeyson2 (16:11 20/10/2003)
          Simo (00:17 21/10/2003)
            ksattic (04:36 22/10/2003)
              Steffen Huber (09:41 22/10/2003)
 
JMB Message #92511, posted by jmb at 01:49, 19/10/2003
Member
Posts: 467
Bear in mind that Creator is a 26bit app so the Iyonix figure will be generated by running the test under Aemulor.
  ^[ Log in to reply ]
 
Steffen Huber Message #92512, posted at 20:38, 19/10/2003, in reply to message #92511
Unregistered user To avoid further confusion, I have finally translated the benchmark part of the article into English:

http://www.huber-net.de/iyonixvsomega/bench_e.htm

Real article in English coming soon.

To answer some of the questions: there are indeed situations where the Omega is faster than the IYONIX - reading from and writing to video memory. However, as you can see from the AWViewer benchmark, as long as mainly writing is used, the IYONIX is still much faster overall.

The Creator and PackDir benchmarks run under Aemulor under the IYONIX, so it is unsurprising that they are only "slightly" faster than a Risc PC - but still, they are faster! So for Aemulor-compatible 26bit software, the IYONIX is the first choice.

The current Omega is not using UDMA yet, so the filing system intensive tests will hopefully improve once UDMA is fully implemented - after all, the IYONIX also started without UDMA.

Steffen

  ^[ Log in to reply ]
 
richard cheng Message #92513, posted by richcheng at 15:37, 20/10/2003, in reply to message #92512

Posts: 655
the benchmarks test for stuff users are likely to do, such as redrawing the ArtWorks apple

*raises eyebrow*

;)

  ^[ Log in to reply ]
 
richard cheng Message #92514, posted by richcheng at 15:38, 20/10/2003, in reply to message #92513

Posts: 655
/, dammit
  ^[ Log in to reply ]
 
Phil Mellor Message #92515, posted by monkeyson2 at 16:11, 20/10/2003, in reply to message #92514
monkeyson2Please don't let them make me be a monkey butler

Posts: 12380
Personally, I redraw the ArtWorks apple one or two or five or twelve times a day. Doesn't everybody?
  ^[ Log in to reply ]
 
Simo Message #92516, posted at 00:17, 21/10/2003, in reply to message #92515
Unregistered user Well the RiscPC's are running Select 4.29, so wouldn't Creator be using the inbuilt JPEG routines, explaining why it is so fast compared to Omega (and Iyonix)?

I'm not altogether surprised that the Omega is just a tad faster than a RiscPC, although I wouldn't mind a Kinetic 300 vs Omega (300) comparison.

  ^[ Log in to reply ]
 
Simon Wilson Message #92517, posted by ksattic at 04:36, 22/10/2003, in reply to message #92516
ksattic
Finally, an avatar!

Posts: 1291
I would like to nominate the Artworks apple redraw as the definitive RISC OS performance test. ;)

Come to think of it, I would have to say that I do redraw the Artworks apple probably at least once a week. I have no life.

  ^[ Log in to reply ]
 
Steffen Huber Message #92518, posted at 09:41, 22/10/2003, in reply to message #92517
Unregistered user Simo,

Creator does not use any JPEG stuff from Select. The first version of Select (RO 4.29) does not have the ImageLoader stuff of the later versions, and the JPEG support is exactly the same as that of RISC OS 4.

Anyway, Creator does the conversion on its own - see its manual.

I am also not surprised that the Omega is only slightly faster than the Risc PC in most benchmarks - this was to be expected. However, I was very surprised to see the huge performance impact of the used screen resolution, and the fact that the Omega manages to be slower in some benchmarks than the Risc PC.

Steffen

  ^[ Log in to reply ]
 

The Icon Bar: News and features: Omega and Iyonix compared