Thursday, December 20, 2007

Electricity Situation Gets Worse

Eskom asked for an 18% price increase, and has been given 14%. So much for the government protecting the consumer or caring about business.
How we have to put up with "load shedding" (i.e the power goes off for 2 hours in randomy selected suburbs) and free electricity to Zimbabwe as well as higher prices and general technical incompetence. Fires at generating plants and substations are usually caused by poor maintenance or obsolete equipment. One fire resulted in the loss of 500MW generating capacity. Then there was the Koeberg debacle which saw rolling blackouts happening for months.
How can this problem be solved? Some people blame it on bad decisions, others blame it on affirmative action leading to the employment of inexperienced engineers. I say that whatever the cause, our current leadership will blame it on Apartheid and do nothing else. It's a failure of power, plain and simple. That's why I think it is appropriate to drop the "e" from their logo. It better reflects what the public feels about them shedding their responsibilities.

Tuesday, December 18, 2007

Charismatic Dimwit

It seems to me that our country is generally run by a*s*holes, and this has been going on for decades, with one notable exception: Nelson Mandela. All the other presidents in my lifetime have been a national embarrassment. I include in this lineup notable disasters like BJ Vorster, PW Botha (the absolute worst that Africa can offer), FW de Klerk and Thabo Mbeki.
If PW and FW had their way, I would have spent time in gaol. So much for them and their "volk and vaderland" which never included white South Africans who didn't speak Afrikaans, and it was always debatable how many Afrikaners it included anyway. I still have difficulty singing "Die Stem" as part of the national anthem, but it's something I'll get over.
Mandela was a great president, even if he did allow thousands to die of AIDS while he was President, and largely ignored the question of crime. At least he did something about AIDS later, hence 46664. His successor, Thabo Mbeki, made all the right noises, changed the way government is structured, but never got a grip on the real issues. Instead he surrounded himself with incompetent dolts like the Minister of Health, who succeeded in screwing up the health system pretty successfully. I always thought she was just holding out for a bribe, but it turned out to be far worse.
So when the ABM movement started (this time it stood for Anything But Mbeki, not Anything But Microsoft) it wasn't because of outrage over health or electricity, but because a charismatic dimwit wasn't properly prosecuted and was allowed to engage in unprotected sex while being the head of the "Moral Regeneration Movement" [cough!]. Or something. As someone once said, "the rabble wants to be roused".
I guess poor people are sick of being given promises that aren't met. So they elected someone who made even bigger promises, while building a big mansion and accepting financial contributions. Go figure. I guess this just proves what American politics has taught us for some time: democracy is where the majority gets what the majority deserves. America got 8 years of George Bush. We've got Mbeki and now we're getting Zuma. Nothing changes, really.

Friday, December 14, 2007

Benchmarks*: Puran Defrag 4.0

Puran Defrag 4.0 was released recently. You can find a review of version 3, and the test results are shown below. It's results are impressive, and much better than I expected, and better than version 3.
The graph shows Puran Defrag 4 (PD4) in dark blue, and the results from Windows XP in light blue. The first test is at the bottom, and shorter lines mean faster times.
  • "Basic XP" refers to the standard install, without Office 2007, so there are only 710 files to be tested.
  • "Defrag" refers to the read times of the same 710 files, after two defrag passes using the full PD4 boot time defrag, with no "Optimize Files" or "Optimize Boot" options enabled. A 17.3% performance improvement is measured.
  • "Defrag+Auto" refers to the read time after enabling both the "Optimize Boot" function in Windows, and then allowing Windows to optimise the placement of system files. After this a further boot time defrag is done, with no additional options. An improvement of 32.6% is recorded, 8.1% faster than WDD.
  • "XP+Auto" is the result obtained after enabling both "Optimize Files" and "Optimize Boot" functions in Windows, and then running WDD, i.e. the best that Windows XP can manage.
  • "Basic Office" refers to the read time of all 802 test files, where no defragmentation has been done whatsoever, after the installation of Microsoft Office 2007 Professional (Trial).
  • "Full Defrag" refers to the read times of the same 802 files, after a full boot time defrag with all options enabled, with no "Optimize Files" or "Optimize Boot" options enabled. A 27.6% performance improvement is measured, as good as the Office Auto result below..
  • "Full Defrag+Auto" refers to the read time after enabling both the "Optimize Boot" function in Windows, and then allowing Windows to move the system files around. After that a simple boot time defrag, with no additional options, was run. An improvement of 32.3% is recorded, 6.2% faster than WDD.
  • "Office Auto" is the result obtained by allowing Windows XP to do its own defrag after enabling both "Optimize Files" and "Optimize Boot" functions.
The performance improvement ranges between 17% and 32%, with an improvement when the boot optimize information is used.
An average 27.4% performance improvement over no defrag at all is excellent for a $24.95/$14.95 (special offer) commercial package, beating all the other "heavy hitter" commercial programs tested so far. In most cases it works as well as or better than the built-in windows defrag. It also has an optional automatic defrag that is really not intrusive at all, and some other useful features. The only thing "missing" is a file placement display. The picture at the top of this article is a composite of two "Analyse" screen shots, one taken after installing Office 2007, and the other taken after the final test, effectively showing before and after situations.
The next test candidate is UltraDefrag 1.2.3

Thursday, December 13, 2007

Puran Defrag 4.0 special offer

The guys at Puran Software are celebrating the release of Puran Defrag 4.0 by dropping the price for a limited time: US$ 14.95, a saving of $10 on the regular price. Even at the higher price it's worth it; this is a bargain!
I have been testing the new version for two weeks, and its brilliant. The boot time defrag goes a good job of tidying up the drive and consolidating free space, and the automatic defrag works pretty well. There are a bunch of small features that I like, such as the ability to do a defrag of a particular file or folder by right-clicking on it. Also, you can do a complete defrag followed by a shutdown, which is useful in an office environment. Just start the defrag and go home. It even runs chkdsk first, as a safety measure. This product is well engineered and the help file is clear and informative.
I plan to publish the benchmarks shortly: most of the delays are out of the way now. I suspect that the price drop has something to do with the outstanding success of their cricket team, but I'm only guessing. ;-)
Update 1 Jan 2008: the special offer price is now $19.95

Benchmarks*: Norton Utilities Speed Disk 19.0.1.8

It's sad to see an old faithful program come to such a tragic end. Version 19 of Speed Disk is part of a humongous bloatware package from Symantec that requires around 300MB of free space just for the install files, unless you do the install directly from the CD. I tried the trial version of the Basic Edition of Norton SystemWorks (ironic, the name) that costs $49.95, and includes Norton Antivirus (NAV) and a whole load of other stuff. I avoided installing the NAV software because I didn't want it to slow the system down. It has been known to turn perfectly respectable machines into clunkers in a matter of minutes. Even the budget version of this package is more expensive than Diskeeper Professional. Sadly, it offers very little in terms of performance. The final defrag test ended up being slower than no defrag at all, and the average improvement was only 4% faster than no defrag, hardly justifying the expense.
The graph shows Norton Speed Disk (NSD) in dark blue, and the results from Windows XP in light blue. The first test is at the bottom, and shorter lines mean faster times.
  • "Basic XP" refers to the standard install, without Office 2007, so there are only 710 files to be tested.
  • "Defrag" refers to the read times of the same 710 files, after two defrag passes using NSD, with no "Optimize Files" or "Optimize Boot" options enabled. No performance improvement is measured at all.
  • "Auto+Layout.ini" refers to the read time after enabling the "Optimize Boot" function in Windows, and then running NSD after Windows had done is standard boot optimise. An improvement of 14.1% is recorded, still slower than WDD.
  • "XP+Auto" is the result obtained after enabling both "Optimize Files" and "Optimize Boot" functions in Windows, and then running WDD, i.e. the best that Windows XP can manage.
  • "Basic Office" refers to the read time of all 802 test files, where no defragmentation has been done whatsoever, after the installation of Microsoft Office 2007 Professional (Trial).
  • "Full Defrag" refers to the read times of the same 802 files, after two defrag passes using NSD, with no "Optimize Files" or "Optimize Boot" options enabled. A 17.2% performance improvement is measured, the best result obtained.
  • "Full+Layout.ini" refers to the read time after enabling the "Optimize Boot" function in Windows, and then running NSD several times. An improvement of -14.2% is recorded, much slower than WDD.
  • "Office Auto" is the result obtained by allowing Windows XP to do its own defrag after enabling both "Optimize Files" and "Optimize Boot" functions.
The performance improvement ranges between -14% and 17%, with mixed improvement when the layout.ini information is used. NSD has no way of enabling or disabling this feature in Windows. Judging by the drive image at the top of this entry, the program is unable to move numerous files (notice all the grey blocks at the bottom), and the pink area at the top of the image is supposed to be the MFT reserved space, but isn't. It's a mess, and it takes forever to run the defrag.
An average 4.3% performance improvement over no defrag at all is extremely disappointing for a $49.99 commercial package, lagging way behind all other programs tested. The program seems to be unaware of the built-in Windows optimisation, or the placement of the MFT reserved space. Frankly, it's a disaster and an embarrassment to Symantec. Judging by the marketing information associated with Norton SystemWorks they already know this, and refer to defragmentation only in passing. Sad.

Sunday, December 02, 2007

Benchmarks*: Disktrix UltimateDefrag 1.72

I have reviewed a previous version of UltimateDefrag 1.72, and posted the results of testing here. This has been the most complex of all the products to benchmark, because of the bewildering array of file placement options. The results show a representative cross-section of these options, using default values wherever possible. I have no explanation why the basic WDD results are faster, but these are still raw results.
The graph shows UltimateDefrag 1.72 (UD) in dark blue, and the results from Windows XP in light blue. The first test is at the bottom, and shorter lines mean faster times. The magenta lines are the results obtained using each of the main file placement options.
  • "Basic XP" refers to the standard install, without Office 2007, so there are only 710 files to be tested.
  • "Defrag Only" refers to the read times of the same 710 files, after two defrag passes using the UD "defragment only" option, with no "Optimize Files" or "Optimize Boot" options enabled. A 17.3% performance improvement is measured.
  • "Consolidate" refers to the UD "Consolidate" option, which returned a 3.1% performance improvement.
  • "File Sort" refers to the UD "File/Folder" option, which returned a 15.1% performance improvement.
  • "Recency" refers to the UD "Recency" (Date Modified) option, which returned a 13.9% performance improvement.
  • "Volatility" refers to the UD "Volatility" option, which returned a 11.5% performance improvement.
  • "Auto" refers to the UD "Auto" option, which returned a 16.7% performance improvement.
  • "Auto+Layout.ini" refers to the read time after enabling both the "Optimize Boot" function in Windows, and then running UD Auto with "respect Layout.ini" enabled. An improvement of 9.1% is recorded, still slower than WDD.
  • "XP+Auto" is the result obtained after enabling both "Optimize Files" and "Optimize Boot" functions in Windows, and then running WDD, i.e. the best that Windows XP can manage.
  • "Basic Office" refers to the read time of all 802 test files, where no defragmentation has been done whatsoever, after the installation of Microsoft Office 2007 Professional (Trial).
  • "Full Defrag Only" refers to the read times of the same 802 files, after two defrag passes using the UD "defragment only" option, with no "Optimize Files" or "Optimize Boot" options enabled. A 13.6% performance improvement is measured.
  • "Consolidate" refers to the UD "Consolidate" option, which returned a 18.3% performance improvement.
  • "File Sort" refers to the UD "File/Folder" option, which returned a 16.5% performance improvement.
  • "Recent Access" refers to the UD "Recency" (Date Last Accessed) option, which returned a 21.1% performance improvement.
  • "Recent Modified" refers to the UD "Recency" (Date Modified) option, which returned a 22.7% performance improvement.
  • "Recent Created" refers to the UD "Recency" (Date Created) option, which returned a 16.6% performance improvement.
  • "Volatility" refers to the UD "Volatility" option, which returned a 23.2% performance improvement.
  • "Auto" refers to the UD "Auto" option, which returned a 20.1% performance improvement.
  • "Auto+Layout.ini" refers to the read time after enabling both the "Optimize Boot" function in Windows, and then running UD Auto with "respect Layout.ini" enabled. An improvement of 21.0% is recorded, still slower than WDD.
  • "Office Auto" is the result obtained by allowing Windows XP to do its own defrag after enabling both "Optimize Files" and "Optimize Boot" functions.
The performance improvement ranges between 3% and 23%, with no noticeable improvement when the layout.ini information is used.
An average 16.2% performance improvement over no defrag at all is a little disappointing for a $39.95 commercial package, lagging behind the other "heavy hitter" commercial programs. Still, dedicated performance junkies can tweak the myriad options to their hearts content until they get the results they desire.
The image here is for the second "Volatility" option. Notice how the file placement is completely different to the "Auto" option at the top of the article.
The next package to be tested is Puran Defrag 4.0, which has just been released.