Pitting a Mac Plus Against an AMD Dual Core

In the Age of Computer Bloat someone has decided to do a performance comparison between a 1986 Mac Plus and a 2007 AMD Dual core, each running appropriate software. Computer Bloat does not fare so well. "In order to keep the hoots and hollers of 'unfair comparison' at a minimum, we designed the tests to be as fair and equitable as possible. We focussed on running tests that reflect how the user perceives the computing experience... And no, we didn't include processing-heavy modern software like Photoshop or Crysis! We selected very basic everyday functions that were performed equally by the 1980's and the 2007 Microsoft applications."

Comments

Tiny midget wizard.

[spun]

He likes ice cream. Spoon it right into the "cup holder" and he might grant you a wish!

Re:Tiny midget wizard.

[spun]

Oh I don't care. Someone saw it, and laughed, that's all I care about.

Developer motivation

[chriss]

I cannot really agree with these tests that just compare "start up tasks" like opening a file or booting the OS. There often is a good reason not to focus too much on these events, because don't happen that often. Responsiveness during use is a better comparison, and this is much harder. Modern machines do a lot of things in the background, like running full blown TCP/IP stacks, something the Mac Plus could not have done. And while opening a file 0.2 seconds faster will not really improve my productivity by much, having instant access to Google and Wikipedia does.

But anyway: Here is a quote from Andy Hertzfeld about how Steve Jobs motivated them to make the Mac boot faster (taken from the documentary The triumph of the nerds by Robert X. Cringley.)

Steve was upset that the Mac took too long to boot to boot up when you first turned it on so he tried motivating Larry Kenyon by telling him well you know how many millions of people are going to buy this machine - it's going to be millions of people and let's imagine that you can make it boot five seconds faster well that's five seconds times a million every day that's fifty lifetimes, if you can shave five seconds off that you're saving fifty lives. And so it was a nice way of thinking about it, and we did get it to go faster.

Re:Developer motivation

[Kadin2048]

I admit the boot-time figure isn't anything to obsess too hard about, but things like application-launch times certainly are. How quickly an application launches adds a lot to how often I use it and how reluctant I am to open it. If I know that launching it is going to take a minute or two (like Photoshop used to on my old PowerMac), I'm not going to click that sucker without a damn good reason. In fact I'm probably going to find some other tool to do the job, if I have a lot of quick tasks to accomplish.

Similarly, if an app takes a long time to save a document, and it blocks the user from doing other things during this process, that's pretty obnoxious. Most people save frequently (or at least they should), and if it takes longer than a second or two at most, you've just interrupted their workflow.

UI responsiveness is definitely king, I'm firmly with you there, but speed in other areas shouldn't just be written off. Applications and system software needs to be designed to do what the user wants, while getting in the way as little as possible. Sometimes I think that gets forgotten by developers, from time to time.

Re:Developer motivation

[l4m3z0r]

I cannot really agree with these tests that just compare "start up tasks" like opening a file or booting the OS. There often is a good reason not to focus too much on these events, because don't happen that often.

I have no idea what the hell your talking about, I open hundreds of files a day on average, and very likely thousands, any programmer working on a large project opens countless files all day long.

Re:Developer motivation

[Anonymous Coward]

Modern machines do a lot of things in the background, like running full blown TCP/IP stacks, something the Mac Plus could not have done.

Oh really?

TCP/IP is not exactly a complicated set of protocols. Ancient machines can and did easily handle it.

Re:Developer motivation

[griffjon]

compare "start up tasks" like opening a file or booting the OS. There often is a good reason not to focus too much on these events, because don't happen that often.

Based on this post, I believe that you must use a Mac, and are just defending the poor 1986 mac.

You don't open files often, so you're not a Linux user. Those guys open files like crazy, all the time. Like, everything is a file to them, and then they open it.

You don't reboot often, so you're obviously not a Windows user.

Please be clear and reveal your personal biases in such important benchmark test discussions.

Re:Developer motivation

[vought]

I wonder how well a modern UNIX would compare to the old Mac

A/UX booted in about 90 seconds on a Mac Quadra 950. 33MHz 68040. As far as I can tell, it had most of the functionality of Mac OS X today - albeit without the slick visual effects and more modern GUI.

Re:Developer motivation

[digitalderbs]

if you can shave five seconds off that you're saving fifty lives

then getting people to ditch their computers completely is like curing cancer and AIDS.

Re:Developer motivation

[gurps_npc]

You misunderstand the point. Basically, the author is saying what you describe is a BAD BAD IDEA.

The writers know about the issue you are talking about and believe that all the crap that they have a modern computer load is NOT neccessary.

Me personally, I know that EVERYTIME I install software, no matter how rarely I wish to use it, I have to check and remove all this GARBAGE that they put into my start up. You gave a list of things such as scanners, DVD burners. I use those rarely.

For 99 out of 100 people there is NO good reason to put those things in the startup. Those are great examples, proving my point. It makes far more sense to 'start' those processes once a month when you actually use them instead of taking 1 second every single day.

If you personally use them every day instead of 1/month, then fine YOU can put them in your startup. Wasting my time (and worse, using vile, hard to understand names making it dificult to realize what your PC is doing and therefore hard/dangerous to remove) placing all that CRAPWARE into startup is obnoxious, rude, and bad business

Re:Developer motivation

[tyme]

griffjon wrote:

You don't open files often, so you're not a Linux user.

so, clearly, I cannot choose the cup in front of you.

You don't reboot often, so you're obviously not a Windows user.

so, clearly, I cannot choose the cup in front of me.

Re:Developer motivation

[encoderer]

Well, to be pedantic, ancient machines were running Trans-Cave Protocol / Abacus Protocol. TCP/AP proved wildly successful. It's affects on the economy were downright chiseling.

Re:Developer motivation

[default luser]

The launch time is a test the modern computer is guaranteed to lose. This is because the peak read rate of hard disks has only improved by 100x in the last 20 years.

Let's start with the admission that the modern OS is 15 thousand times larger (1MB versus 15GB). It's a fair assumption that most applications are at least larger by a small fraction of that - say one thousand times larger.

That old SCSI hard disk would have a peak read speed of around 1MB/s, while the best disks around today are approaching 100MB/s. That's a 100x increase in read delay speed, ignoring access time improvements.

Even if we're generous and say the application code being loaded is a slim one thousand times larger than the version on the Mac Plus, with the pitiful 100x increase in hard disk read speed, it's still going to load 10x slower.

So, why are they so close? Why is the modern system with the relatively slow hard disk so competitive?

* Fast caching architecture with prefetch (on the hard disk controller itself).
* Shared libraries are already memory-resident.
* OS attempts to preload your most-used applications into memory (new feature in Vista).

All these features are designed to get around the limitations imposed by the hard disk. Once we finally find a better solid-state solution, the modern system could potentialy blow the Mac Plus out of the water.

Spoiler Warning

[Average_Joe_Sixpack]

*** the webserver lost

Re:Spoiler Warning

[missing000]

But is it running on a Mac+ or AMD?

Lets compare a typewriter to a word processor.

[TheLazySci-FiAuthor]

As with this article, what really would that prove?

I think I would find my words per minute would not vary. The legibility of the document would be identical. I could even say that the typewriter is superior in some ways - for instance, my document autosaves on every keystroke.

Calling features "bloat" strikes me the same as when a person will call a reason an "excuse". There are times and places when "bloat" and "excuses" are valid words, but they can be inserted where they are invalid just as easily.

Perhaps the law of diminishing returns holds true. After all, a typewriter really is all one needs to write a novel, and in fact I do not think a computer helps one write a novel thousands of times more quickly. However, there are features (spell check, formatting, fonts, predictive text, voice recognition...) that enhance the writing experience.

I guess I just don't get the point of this article.

Re:Lets compare a typewriter to a word processor.

[rubycodez]

actually, you're half right on the features: spell checking, formatting and fonts are useful and can be done with the 20+ year old software as well as the new (and faster on the old). But to anyone who types at a reasonable rate, predictive text is a huge annoyance (and often causes wrong values to be input into fields and just slows the typing of documents. Good typists turn that crap off, it IS bloat. Voice recognition is much slower and much less accurate than typing, I wouldn't even consider using it to create a document. But bloat and gee-whizz panders to the "hunt and peck" crowd.

Re:Lets compare a typewriter to a word processor.

[jfengel]

Having written a few books myself, I can say that the biggest advantage of a computer over a typewriter is the ability to correct and reorganize.

I once reviewed a preliminary copy of a text by Jef Raskin, one of the Mac designers. It was double-spaced Courier, with hand-drawn diagrams. I found that ironic, coming from him, but it made sense. There were professionals to make the drawings look nice and format the text. His job was the words and the gist of the diagrams.

Nonetheless, it was typed on a computer. (It's easy to spot typewritten text; it will always have some typos or irregular letters). I'm sure it's because it let him rearrange sentences, paragraphs, and even chapters without having to re-type from scratch, and it's no harder than typing. The diagrams, however, are still more work than hand-drawing. (At least, I know of no tool that's as easy, even with a drawing tablet.)

Some writers prefer the notion of organizing everything in your head before typing anything, but that's more memory than I've got. I relied on the ability of the word processor so I could start a paragraph and come back to it later without having to change the paper in my typewriter a huge time boost.

Despite what I've just said, I concur that the article is mostly silly. Others are making that point as well as I can. I just wanted to show why I thought a computer was much better than a typewriter, for different reasons than you gave.

Re:Lets compare a typewriter to a word processor.

[phasm42]

spell checking, formatting and fonts are useful and can be done with the 20+ year old software as well as the new (and faster on the old).

What if you need to work in Swedish and Japanese documents? Oh wait, only English matters.

But to anyone who types at a reasonable rate, predictive text is a huge annoyance (and often causes wrong values to be input into fields and just slows the typing of documents. Good typists turn that crap off, it IS bloat. Voice recognition is much slower and much less accurate than typing, I wouldn't even consider using it to create a document.

I guess disabled people shouldn't be using computers then?

Re:Lets compare a typewriter to a word processor.

[phasm42]

Who said anything about loading it into RAM? But the mere ability to support these things does increases the complexity of a program greatly, and thus increases resource usage.

Re:Lets compare a typewriter to a word processor.

[rubycodez]

There in fact were computerized solutions 20+ years ago for writing documents with Scandinavian characters and Japanese pictographs/romanizations that are less bloated than today's.

A disabled person may find themselves with inferior production tools, and that is what the current state of the art gives them if they use voice recognition compared to someone using typing. Meanwhile, for the 99.999+% of the human race with fully functioning fingers, they'll do better to learn to type properly.

"I have no substantive arguments on the subject so I'll try to invoke guilt of the plight of the less fortunate or guilt of racism because needs of ethnic group x wasn't addressed".

Re:Lets compare a typewriter to a word processor.

[value_added]

Some writers prefer the notion of organizing everything in your head before typing anything, but that's more memory than I've got. I relied on the ability of the word processor so I could start a paragraph and come back to it later without having to change the paper in my typewriter a huge time boost.

Years ago when I worked for a large law firm, I never ceased to be amazed how the "old timers" (partners who grew up in the days when things like secretarial pools existed) could and regularly would dictate their work in full; the secretary would type it up and that, with the possible addition of some whiteout, was the end of that. By contrast, most of the newer attorneys and paralegals would typically do their work entirely on a computer. A typical transmittal letter ("Please find enclosed blah blah") would require 20 or more minutes of work, subject to any number of further revisions inspired by each successive printing, supervisory review of redlined markups, and repeated spell czechs. To be fair, in that world, every word and punctuation mark takes on significance, but you get the idea.

Similarly, consider that great novels didn't come into existence with the advent of word processing machines. They were laboriously crafted on typewriters, and before that using writing instruments that ranged from ballpoint pens to quills.

My own theory is that we've simply gotten lazy. And the tools we now have at our disposal, while possibly increasing our productivity in certain respects, mostly fill our time and use our energy with superficialities. Put another way, I'm suspicious that the widespread use of wordprocessors has created a state of affairs in which people are "drawing" their documents and stuffing them with bits of disjointed thoughts (to be revised later) instead of actually writing the damned things.

The ability to revise a document on a computer is marvelous, I'll agree, but what if that ability compensates, like 500 monkeys with typewriters, for an innability to think, and by extension, to write? And if that's not the case, how then to explain the current state of general illiteracy (as evidenced by an innability to spell or distinguish certain words) in a population where the use of and reliance on automated spell checking is so widespread?

Garrison Keillor once said that it's better to write "three sharp and funny pages about geese than 300 fat and flabby ones about the human condition." I doubt he used a word processor to come up with that one. Not everyone has a talent for writing, of course. But I suspect that most of those that don't are spending so much of their time adjusting fonts and paragraph styles in their 300 page Word version of the human condition to know how fat and flabby things really are.

Huh?

[phasm42]

Why didn't he compare the Mac Plus against an OS X machine, or the XP machine against a DOS 6 machine?

Also nice how everything that the Mac Plus (and old machines in general) sucked at or couldn't do were left out. Making such a big deal out of startup time seems pretty pointless too.

Re:Huh?

[chriss]

Why didn't he compare the Mac Plus against an OS X machine, or the XP machine against a DOS 6 machine?

Because the Mac Plus and the WinXP Pro SP2 systems were the most widely used GUI based desktop machines at their respective time, thus making a comparison about productivity feasible.

Re:Huh?

[jellomizer]

Well the Mac Plus was a GUI system at the time so it is a closer match for comparing Windows XP then DOS 6 is. As well many of the tests used would be unfair with OS X for Intel. Excel and Word for mac are still Power PC version so they need to run via Rosetta which will slow down the resusults running XP will actually have a better performance for the Test.

Yes everyting the Mac Plus couldn't do was left out. Also they didn't run the normal benchmark software as well. Knowing quite well the new system will eat its lunch. Also they are using different versions of software. But the point of the test was comparing the quality of life for people with the Mac Pro back in the 80s vs. the Quality of Life today with people with PCs today, doing the same jobs.

Bootup Speet is important espectially back in the 80's where people turned off their computers when they were done, and people still do that today. So bootup time is quite useful in measuring productivity. In Linux if you misconfigure say sendmail in Red Hat when you boot up you are waiting for minutes for it to load and fail. Making Linux Boot time painfully slow. This effects productivity (say your job is to insure Sendmail works properly at bootup). For windows reboots are frequent when you have updates so you are working on you job and you get an automatic update you need to reboot and wait 2 minutes when you get everything back you need to refresh were you left off.

The point of the article is that as computers get faster the software get proportionally slower so you tend to get a 0 net gain in productivity in the common jobs you do on your system now.

Re:Huh?

[jellomizer]

Not really, when it's a cost you pay once a day, which you can spend doing something other than staring at the machine.
Unless you you have a reboot in the middle of the day from a power outage, Update, System Crash, or you are just late and you need to get a file off your system which was powered down.

You don't need to reboot a machine to configure sendmail. And who has a job watching sendmail boot? This doesn't make any sense.
Unless you want to be sure it goes back up properly after bootup. Say you were altering the S99Sendmail file for different options and while you tested it and it workes fine without the reboot. When a reboot did happen (long power outage, Kernel Update, New hardware/Hardware failure) it fails to load properly because say you had it installed for the wrong init level. When configurating a system I like to prove that it will come back on in the state I want it to be in.

2 minutes, once or twice a month? I think that's negligible for a desktop machine.
It could also be Once a day to many times a day. Say you have a Nazi Security Adminsitratior who for the last couple of weeks has been perfecting his security policy and forces a reboot after every attempt. So 2 or 3 times a day gets annoying and waiting for it to reboot for a couple of minutes makes it worse.

If you ignore all the jobs you couldn't even do on the old system.
Yes but a lot of people/companies upgrade weither they need it or not. So they will get a new system and gain little, for most cases it will not be as big as a Mac Plus and a New PC 20 years later. But lets say from a Windows 98 System a P2 300Mhz to a Core 2 Duo 2.66 ghz running Vista (10 years (More or Less) difference). With Office 97 Upgraded to Office 2007... The bulk of the work you will do yesterday vs. today on the new system would be the same and substracting any learning curve your net productivity will be 0.

The Maytag of Computers

[Spencerian]

The Mac Plus (of which I was a former owner) is a quintessential example of Apple's past design principles in terms of quality (recent examples such as the Macbooks, which I also own, are having nasty hardware and QC issues). The fact that you can get this old Mac to speak "internet" and continue to run (it has only a SCSI-25 interface for drives and other peripherals) is a testament to good design, whether you're an Apple fan or not.

Finding a contemporary IBM PC to do the same performance test would be more appropriate and interesting, but connectivity and functionality there (it was built years before Windows) would be a big challenge under the non-graphical DOS, if not impossible. I don't know if there's even a Linux out there that could understand that old PC technology. I'm sure it could be done--I just wouldn't want to be the one to try.

1986 software at least felt deterministic

[ribuck]

Software from 1986 didn't have scalable fonts, 32-bit colour, etc, but the interface was usually snappy. Menus dropped down snappily, and dialog boxes opened immediately, for example.

Operations that took a long time (such as reflowing a page in a desktop publishing program) at least appeared deterministic - you knew it would take a second or two to reflow, so you weren't anxiously waiting for the system to do something.

Slashdotted!

[rueger]

Guess we've established that the Mac Plus was not the best choice for hosting the web site?

glad someone did this comparison...

[sloth jr]

I've been troubled for years on how generational improvements in computation equipment don't seem to result in improved USER experience. Now, important to realize that in the comparisons selected, we're talking about 1-bit bit-mapped operations on a screen 512x368 in size (from memory - might have botched the Y coord limit). Might be interesting to see what happens on that PC when dropping the display to 640x480 and 256 colors. That'd be a little closer to apples-apples comparison.

I digress. The point is - nothing seems much better in the user experience than before, for the vast majority of things we do - and that includes MacOS X, to my thinking. Nothing that makes me jump up and down and twist and shout anyway. What apps have I added in the last 10 years? Music players. Video players. Browsers. Pretty much it. I wonder where the hell my 4.5 billion clock cycles a second are actually going.

I don't know - computing just doesn't seem very exciting anymore. Help.

sloth jr

Re:glad someone did this comparison...

[Erwos]

"The point is - nothing seems much better in the user experience than before, for the vast majority of things we do - and that includes MacOS X, to my thinking."

Yeah, except for multi-tasking. You don't use that at all, right? Multifinder was only introduced in System 6, long after the Mac Plus was made obsolete by newer Macs.

Look, I liked my Mac Plus. I even liked the 512k, except for bumping against memory limits in large documents. But you're really viewing this whole thing with rose-colored glasses if you think that a Mac Plus with a bigger screen is going to be just as usable or pleasant-to-use as Vista or MacOS X.

Re:glad someone did this comparison...

[*weasel]

I've been troubled for years on how generational improvements in computation equipment don't seem to result in improved USER experience.

Improved USER experience almost always comes from new software/features, rather than improvements to old software/features. The new features are where your clock cycles go. It's where they've always gone.

Word isn't opening any faster twenty years on. But it is spell-/grammar-checking the document, importing multimedia, rendering a cleartype font, looking for online collaborators, refreshing relevant Research tab tools, etc.

Those tools wouldn't be possible, in-line, on-demand, without those few billion new clock cycles. It'd be nice if things worked faster, and they undoubtedly could. But let's not short-change the user experience advancements that all the features they enabled.

Also, let's not gloss over browsers and email. Those two have resulted in absurd quantities of user enrichment.

Who cares if Word doesn't open any faster if John Doe has instant access to every reference material known to man? What saves him more time: a document opening a half-second sooner? Or less time spent dealing with interoffice mail, less phone interruptions and less hunting down, storing, sorting and searching through physical reference volumes?

your personal machine might not be making the best use of all those cycles, but that doesn't mean your user experience isn't being enriched by them.

Comparison to AMD

[phasm42]

The constant blather about comparing it to "AMD" really speaks volumes about the author. Apparently AMD determines your user experience on a modern PC running XP.

Oh, and browsing the web plays no part in the modern user experience. None at all. Don't even think about it. If most people weren't doing it in '86, it's not important.

Which is why I like vi...

[wandazulu]

I used vi on a VT100 attached to a vax running BSD back in 1990 and I use vi (vim) today on a MacBook Pro that could handle more simultaneous users than the vax did. It was always fast to start then, and it's fast to start today, though now I have colors, split windows, and a bajillion other features I struggle to remember.

It's interesting to see that the machines have gotten faster, software more complex, etc., etc., but software like vim just keeps on truckin'. Too bad we don't have more software like this.

Re:Which is why I like vi...

[timeOday]

I used vi on a VT100 attached to a vax running BSD back in 1990 and I use vi (vim) today on a MacBook Pro that could handle more simultaneous users than the vax did. It was always fast to start then, and it's fast to start today

How the heck did that get modded flaimbait?

Me, I use emacs. It was sluggish back then, and it's still sluggish now :) (At least, the startup time is a little annoying).

Also

[Sycraft-fu]

It ignores other factors such as relative price. Of course the server is apparently being run on a Mac plus so I can't go back and check the article to see if they listed the specs of the current PC, but a Mac Plus cost about $2500 when it was introduced. Now, take $2500 in 1986 dollars and you get about $4500 in today's dollars. Well, $4500 buys you a shitload of computer. You can get a much better processor than they had, 4GB of RAM, a hardware RAID controller with a bunch of disks and so on. Load up something like that and see what your launch times are like. Given that the system they are using probably is less than $2000 in today's dollars, you aren't even close money wise.

However as you said, it doesn't really matter as the computers are performing on totally different levels. In every way a new system does more than a Mac Plus. Even if you dismiss the usefulness of multi-tasking and look at just the app there's huge improvements. One would be the in-line spell checker. As I'm sure this post is revealing I'm a horrible speller. However in Word it is great, it will check spelling as I go along. After a few times of correcting the same mistake, it just starts auto correcting. It gets to the point where once I've trained a copy I can type a document and it is good to go as it has fixed all the problems.

This is just another example of the great Benjamin Disraeli quote: "There are three kinds of lies: lies, damned lies, and statistics." Or in other words, you can twist around a test in almost any way you like to make it come out with a result that you want. However that doesn't mean that it has any relevance.

Bah

[Stickerboy]

Only the AMD dual core can run Clippy while you go about your work, that's the only thing that matters in my book...

Maybe read the whole article?

[LoaTao]

"Is this to say that the Mac Plus is a better computer than the AMD? Of course not. The technological advancements of 21 years have placed modern PCs in a completely different league of varied capacities. But the "User Experience" has not changed much in two decades.' The point they are trying to make is that while hardware has advanced considerably and software has added capability the actually base user interface and thus user experience has not changed much in 20 years. The authors don't go into the "why" and do show some personal bias and opinion... but they have a point... and, yes, I still own a functioning MacPlus.

The Anthropic Principle

[briglass]

The anthropic principle helps to explain why this comparison makes no sense. By virtue of the fact that both computers are market-ready and market-tested machines (especially in the highly successful Mac Plus), their usability speeds MUST be under or around market-acceptable levels. Otherwise, they would either not have survived alpha and beta testing or not have survived as a marketable product. What this comparison is really tapping into is the user-acceptable speed level, which has not changed since the 1980s (because humans haven't changed much).

First Post!

[Timesprout]

I submitted this from my 1985 Amiga to proved that even a 1986 Mac is bloated and slow.

Re:First Post!

[the_greywolf]

Compared to an Amiga, everything is bloated and slow.

... and when do I get fast copper on my PC? Hmm?

I knew it..

[GreggBz]

This is proof positive that Steve Jobs traveled to the alternate future in the Dolorean Woz invented to subsequently steal all his technology from the Amiga intergalactic headquarters on Lunar Base Lorannie.

Then one day Bill Gates found Steve's copy of the "Workbench 3.1 Users Guide" sitting in the Delorean, and hijacked it traveling back in time to give his younger self a copy, and therefore, the keys to a multi-billion dollar future of corrupt monopolies.

Re:Why they didn't include linux in the tests

[morgan_greywolf]

You're talking about *rich* cut and paste. The OP was talking about cut and paste. Still works today. Highlight the text, point in the new text input area, click the middle button. Done.

Old PC's w/old OS's can fly.....

[Slugster]

This result (what I can glean from comments, as the site is being pounded) doesn't surprise me.

I have an aging Win98-era Pentium II@350 Mhz with 392 megs of RAM, and running Win98, it simply flies.... I keep it around to run some era programs I like, and every time I power it up, I am simply stunned all over again at how blazingly fast it responds. It responds to user input and opens regular programs noticeably faster than the few computers I've bought since--computers that have faster drives, much faster CPU's and way more RAM. .... I have (single-CPU) WinXP machines (haven't stepped up to any dual-cores yet, but I wonder what good it'll do), have run a couple GUI-distros of Linux on them over the years and have seen Apples at work--and nothing new I've yet seen is as fast as that clunker 98 box is, running 98. :|

Of course Win98 has a number of problems now--a lot of vulnerabilities and no antivirus I know of still supports it, so getting online is walking in a minefield. And even used for local apps it needs to be rebooted every 4-6 hours to be safe... but even then, warm-rebooting only takes like 20 fucking seconds, and that's just the usual OS install, no optimization ever undertaken. Did we used to bitch about bootup times? Have they gotten longer or shorter?

For a whlie I had Mandrake on it too, but Mandrake ran like a dog. With Linux and WinXP there's all this fucking-about with the hard drive that has to occur, for some reason..... any time you do something, even with the hard drives spinning, these bigger/better OS's seemto have to go off and piss away a couple seconds before actually doing anything.

All your boxen belong to bloat.
~

Boiling frogs

[SuperKendall]

...their usability speeds MUST be under or around market-acceptable levels...

"Market acceptable" is a measurement that is not static.

Let's look at the convention wisdom on boiling frogs. Supposedly, if you put them in boiling water they will hop right out - but if you put them in a vat of cold water, they will stay in the pot as you progressively heat it to boiling.

The computer industry has been boiling frogs (where we are all the frogs) for twenty years or so, where the next generation of computers are just a little slower with each iteration. It's not much slower, and offers a bit more, so people accept it - and along with it a new definition of "market acceptable".

So it's not like this article is not raising some really valid points.

Time (not processing) is the constant

[Rui del-Negro]

The real issue here is not if an "AMD system" is faster than a "Mac". For that, they would have to test exactly the same software, not different versions of it. The issue is if modern software, running on modern hardware, is faster than old software, running on old hardware.

For "interactive" tasks it usually isn't, and for a good reason.

No one cares if a program takes 1.4 seconds to complete a find & replace instead of 0.8 seconds. No one cares if a program takes 5.4 seconds to start instead of 3.9. If it took 20, then yes, people probably would care. You see, for interactive tasks, time is the fixed value. Specifically, the time that people don't mind waiting (which varies depending on how common that task is, of course).

This article just proves Murphy's laws of computation: data expands to fill all available space, processing expands to fill all available time, etc..

It's the same thing with games. I could probably take a game from 1995 and run it at 400 fps on my modern hardware. But if I can run a much better-looking version at 60 or even 30 fps, I'll probably pick that one instead. If it ran at 5 fps, I would rather play the old one.

There is a point beyond which "more features" (or "prettier graphics" or whatever) is worth more than an increase in "reaction speed".

That is why CPU-intensive tasks (the ones that never feel "fast enough") are the right way to test hardware; because they tell you how fast the thing can run, and not how fast the developers decided it should run to avoid annoying the user while appealing to as many people as possible (by including extra features).

The article's conclusion that there is "zero advance in productivity" is meaningless. Even if we take one of the most common operations (find & replace), does anyone really believe that, if it completed 1 second faster, people would be noticeably "more productive"...? In this kind of task, "productivity" depends 99% on the human part of the system.

word processor productivity kills typewriter.

[twitter]

Only a person who's never had to use a typewriter could think of it the way they think of a word processor. People dedicated their lives to typing and made careers out of doing it well. The average person gave their hand written manuscripts to secretaries who typed them, if and only if it had to be published. Word processing is much faster, if you have reasonable software. This is why people spent thousands of dollars on computers that did little more than spell check and print.

The authors fairly compare user experience. Things like typing and scrolling lag matter. If you have too much of either, a typewriter might be faster. Of course it takes a lot of lag to make up the time it takes liquid paper to dry.

Did you forget about Liquid Paper? You might have if you used IBM's correctable type ribbons.

You are right about legibility though. OCR can eat typed pages and then your typed manuscript can be modified and duplicated like any other electronic document.

They might call it Computer Bloat...

[The Living Fractal]

I call it being able to surf Wikipedia, Google, and Slashdot in a tabbed browser while running a program like Seti@home in the background with Winamp, Excel, Word and Outlook all readily available at the touch of a button (alt-tab) through mapped servers that centrally store my work. (let's not forget WoW running windowed in the background so I can watch my auctions). Oh, I probably shouldn't leave out the firewalls, the AV software, the synchronization/connection with my PDA, the EPO client, the dual 21" LCDs driven at 1600x1200 EACH, and the fact that all of it pops up on my screen within a second if I want it.

Gee. I guess I don't call that bloat at all. I call it multi-tasking. Let's see a computer from 1986 do that.

So let me get this straight. Someone's complaining that a computer today can do all of this but that dialog boxes pop up a little slower? Then go back to using your '86 Mac. I'm quite happy with what I have today, thank you.

TLF

Eroom's Law

[bflynn]

Eroom's Law: Every 18 months, computer software requires twice the computing power to operate.

This is the counter to Moore's law and explains why the testing achieved the results they did.

Brian