dave2002.alt
This is an alternative weblog for the existing dave2002 weblog hosted at dave2002.pitas.com. This features material on weblogs, music, art, theatre, travel, news and software and hardware, amongst other things

My recent furls


Search this site or the web powered by FreeFind

Site search Web search

Site Meter



Monday, September 20, 2004  

I was tempted to try out OnSpeed. I wasn't sure if I needed it, or if it would do any good, as our broadband is generally reasonably good. However since it works by compressing, I figured that it could still make a difference. Initial impressions were favourable, but after a week I've noticed some problems. Mostly they're not major, but for broadband users they're probably sufficient to deter potential users. The package doesn't even pretend to handle streaming data services or music or video downloads, so some heavy users won't get very much benefits really.

My usage to date has resulted in about 66 Mbytes of data being requested, of which 31 Mbytes were actually transferred. This seems significant, and is measured at a ratio of 2.1:1. The measures are probably not completely accurate however, and may ignore data which comes in emails. I haven't been able to get this package working with emails - it doesn't bother me much, but may be a source of problems.

If you consider usage similar to mine the daily data served out is around 12 Mbytes, of which approx 6 MBytes is actually downloaded. That means that roughly 6 Mbytes/day is saved. The time saving across a broadband network running at about 450 Mbps (typical) is about 1 minute 40 seconds per day. That's not earth shattering. Over a year I'd save about 10 hours, assuming I didn't lose it in other ways. The service costs about £25 per year, or approx £2 per month. At this rate I'm spending £25 to save 10 hours, or about £2.50 for each saved hour. Perhaps that's not too bad really, if I can find something more profitable to do with my time, but it's a slow rate of return.

Some of the time saved might be lost in other ways, such as installation, or fixing problems which might arise. One obvious problem is with images, which can be compressed, sometimes very significantly. Small icons and similar images are not a problem, but photos can be degraded quite rapidly with even a small amount of compression. It is possible to force the download of original images, though doing this on a permanent basis annuls a significant part of the speed improvements. It makes more sense to force the download of originals only for pictures which are needed to be of high quality. This can be done by selecting the image, right clicking, and then choosing the option to download the original. Users who spend a lot of time looking at photos might not like this package.

The package does work with Apple machines, such as the iBook, though it may not deliver the same level of performance. The machine I used for testing was slow, and it wasn't improved much by using OnSpeed.

Now consider the effect on dial-up users. Comparisons are probably not completely meaningful, but let's try anyway. Assuming a similar 6 Mbytes saving per day, that works out at 20 minutes or more, and over a year this could result in a saving of 120 hours. That's 5 days! The reason that comparisons are not completely meaningful is that users would have to be pretty dedicated to do this on a daily basis over a dial up, but some users may be doing this. This suggests that dial-up users really could get significant benefits from this software.

I haven't got definite figures yet from anyone who's tried this with dial-up, so I don't know how it really works in practice. Some of the reviews have been really positive, so perhaps this is a good thing for this kind of user.

One other kind of user might obtain significant benefits from this software. Users who use wireless services such as GPRS or 3G, and who are on a plan which charges by the Mbyte could make definite speed and also cost savings. For these users the time saving would be beneficial, but the cost saving would be very much welcomed. The kind of usage I'm getting with my broadband service would be likely to be on a high 3G tariff, so if I were to become wirelessly mobile using cellular networks I'd either be bankrupted, or I'd have to cut back my usage.

Users have up to 14 days to try OnSpeed out. I think it's worth a try - particularly for dial-up and some wireless mobile users. Broadband users will get relatively little benefit - though there is still some effect. There are some disadvantages, and the behaviour is also dependent on software used. OnSpeed works differently with different browsers, and I've not tried them all out.

Some dial-up users use AOL for connections, and AOL also provides speed up technology. AOL users may find that OnSpeed is not compatible with their usage.

The Guardian review suggested that privacy issues were also a factor to be considered, and could put many users off.

An alternative service is Propel - which has been reviewed at rwapadventures.com.

I'm still evaluating the system myself, and trying to work out whether £2.50 per hour for 10 hours saved over a year is really a good deal. Possibly not. Dial-up users with an expectation of perhaps 100 hours saved have an easier decision to make, since 25p per saved hour is not unreasonable. Users on pay per byte wireless tariffs save both on saved time, and also on lower transfer costs.

There are some problems which may be attributable to the software - but mostly it just works!



posted by David | 9:23 PM
Comments: Post a Comment
archives
links