ramdisk

Increasing memory cache to improve Firefox performance

Dave Farquhar Software , , ,

I read today in Lifehacker about disabling Firefox’s disk cache and increasing the memory cache, as an alternative to putting the disk cache on a ramdisk. The trick can work, depending on the types of sites you visit. But the two aren’t quite interchangeable. The disk cache stores compressed images and (I believe) html. The memory cache stores uncompressed pictures for fast rendering, and no html. Content stored in one isn’t necessarily stored in the other. Read more

Using video memory as a ramdisk in Linux

Dave Farquhar Linux , , , ,

An old idea hit me again recently: Why can’t you use the memory that’s sitting unused on your video card (unless you’re playing Doom) as a ramdisk? It turns out you can, just not if you’re using Windows. Some Linux people have been doing <a href=”http://hedera.linuxnews.pl/_news/2002/09/03/_long/1445.html”>it</a> for two years.<p>Where’d I get this loony idea? Commodore, that’s where. It was fairly common practice to use the video RAM dedicated to the C-128’s 80-column display for other purposes when you weren’t using it. As convoluted as PC video memory is, it had nothing on the C-128, where the 80-column video chip was a netherword accessible only via a handful of chip registers. Using the memory for anything else was slow, it was painful, but it was still a lot faster than Commodore’s floppy drives.<p>

So along comes someone on Slashdot, asking about using idle video memory as swap space. I really like the idea on principle: The memory isn’t doing anything, and RAM is at least an order of magnitude faster than disk, so even slow memory is going to give better performance.<p>

The principle goes like this: You use the Linux MTD module and point it at the video card’s memory in the PCI address space. The memory is now a block device, which you can format and put a filesystem on. Format it ext2 (who needs journaling on a ramdisk?), and you’ve got a ramdisk. Format it swap, and you’ve got swap space.<p>

The downside? Reads and writes don’t happen at the same speed with AGP. Since swap space needs to happen quickly both directions, this is a problem. It could work a lot better with older PCI video cards, but those of course are a lot less likely to have a useful amount of memory on them. It would also work a lot better on newer PCIe video cards, but of course if your system is new enough to have a PCIe card, it’s also likely to have huge amounts of system RAM.<p>

The other downside is that CPU usage tends to really jump while accessing the video RAM.<p>

If you happen to have a system that has fast access to its video RAM, there’s no reason not to try using it as swap space. On some systems it seems to work really well. On others it seems to work really poorly.<p>

If it’s too slow for swap space, try it as a ramdisk. Point your browser cache at it, or mount it as /tmp. It’s going to have lower latency than disk, guaranteed. The only question is the throughput. But if it’s handling large numbers of small files, latency matters more than throughput.<p>

And if you’re concerned about the quality of the memory chips on a video card being lower than the quality of the chips used on the motherboard, a concern some people on Slashdot expressed, using that memory as a ramdisk is safer than as a system file. If there’s slight corruption in the memory, the filesystem will report an error. Personally I’m not sure I buy that argument, since GPUs tend to be even more demanding on memory than CPUs are, and the consequences of using second-rate memory on a video card could be worse than just some stray blips on the screen. But if you’re a worry wart, using it for something less important than swap means you’re not risking a system crash by doing it.<p>

If you’re the type who likes to tinker, this could be a way to get some performance at no cost other than your time. Of course if you like to tinker and enjoy this kind of stuff anyway, your time is essentially free.<p>

And if you want to get really crazy, RAID your new ramdisk with a small partition on your hard drive to make it permanent. But that seems a little too out there even for me.

We\’ll have to wait longer for PCI RAMdisks

Dave Farquhar Hardware , , , , ,

In case nobody noticed, it’s August. July came and went, and there’s no Gigabyte I-RAM on the market yet.

But there are a few benchmarks out there, and Anandtech has an article that, once you get past the usual rambling and over-the-top introduction, has some useful insights.I was going to say the first problem is the somewhat disappointing speed, but actually, there are two bigger problems:

Availability. Now they’re saying it’ll be out sometime in August. And they’re initially only going to make 1,000 of them.

Price. The original $50 MSRP is out the window; now this thing is going to cost $150.

Can anything else be wrong? Unfortunately, yes. The speed is a bit disappointing. The SATA interface is the bottleneck. The very newest hard drives can come close to saturating the SATA interface for short periods of time, so the RAMdisk doesn’t outperform it by much. If this drive were using an interface with more bandwidth, there wouldn’t be as much problem, but squeezing more bandwidth out of the 33 MHz PCI bus is tough. We’re at the point now where the PCI bus is a much bigger bottleneck than the ISA bus was in 1994. The theoretical limit of the PCI bus is 132 megabytes per second, which isn’t much higher than the sustained throughput of 100 megabytes per second that the I-RAM delivers.

The combination of PCI Express and a faster disk protocol has the potential to resolve this issue, but at the expense of limiting the device’s market even further.

I’m disappointed by the review in a couple of regards, though. First, they compare the I-RAM to the fastest SATA drive available at the time of the review. That’s not necessarily what every would-be purchaser would be using. I believe that an I-RAM used to replace (or in conjunction with) a drive that’s a couple of years old would be a mind-blowing upgrade.

Second, they don’t take fragmentation into account. Enthusiasts are more likely to defragment their hard drives twice a day than everyone else, so fragmentation may not be an issue for them. But my wife, mother, and mother-in-law don’t know what fragmentation is. Well, maybe my wife does because she’s probably overheard me talk about it. The thing about the I-RAM is that it makes seek times irrelevant, so it’s never going to slow down due to fragmentation. Translation: For people who have lives, this thing could be phenomenal.

The review complained constantly about the drive’s capacity. So I’m disappointed that they didn’t test the drive with NTFS compression enabled. While data compression is still taboo, and it increases CPU usage, when you’re out of room it’s your only choice. While its effectiveness is unpredictable, it’s fairly safe to bet compression will get you another gigabyte or two of usable space on a 4-gig model. But just as importantly, under some circumstances, compression can actually increase performance. I want to know if increasing the amount of data you’re flowing over the saturated bus makes up for the increased CPU usage.

Finally! A $60 RAMdisk on a PCI card

Dave Farquhar Hardware , , , , , , , , ,

PC World: Taiwanese hardware maker Gigabyte Technology has stumbled upon a faster way to boot up PCs based on Microsoft’s Windows XP operating system.

Please allow me to quote something I penned back in 1999: “I’d love to see someone design and release a battery-backed hardware RAM disk for PCs… Such devices existed in the early 1990s for the Commodore 64/128 and the Apple IIgs and permitted these systems to boot their graphical operating systems before the PCs of their day had managed to bring up a C: prompt. A similar device for today’s PCs would do more to boost system performance than any other innovation I see coming down the pipeline any time soon.”You can find the paragraph, in context, on page 214 of Optimizing Windows for Games, Graphics and Multimedia.

Enough self-congratulation. I’m glad someone finally made this device, which is called the Gigabyte i-Ram PCI ramdisk. And here’s the great news: The device is going to cost about $60 without RAM. 512-meg DIMMs can be expensive or cheap. A quick scan turns up some that I’d be willing to trust for $41 from Newegg.com.

It plugs into a PCI slot but it only uses the slot for power. Data itself is transferred via a serial ATA cable. This improves compatibility, I suppose, but I would have liked to have seen the serial ATA hardware integrated onto the board. But that would have increased costs, and arguably most of the people who will want this already have serial ATA. At least the target market does. I don’t know if this is going to prove more popular with people who want to hot rod their Pentium 4s, or people who want to increase the life expectancy of an older PC. This thing would do wonders for Mom’s PC, or my sister’s PC, and their primary interests are word processing and e-mail. They would love the speed and the quiet.

I’ve got all sorts of ideas for this thing. The article says it’ll be out in July. I want one BAD.

What kinds of ideas? For one, I’d love to eliminate the biggest source of latency in my PCs. I tend not to hit the CPU all that hard most of the time, but I sure do hit my disks hard. I’d love to eliminate the last mechanical piece in the system. Let’s face it: Hard drives crash. This thing gets wiped out if it loses power for 12 hours, but how often does that really happen? And if you’ve got a UPS and you shut the system down, shouldn’t it last indefinitely? Backing the data up to a real hard drive on the network somewhere, or onto a memory stick will solve that issue. Between that and a Ghost image of the system partition, you can recover from a power outage fast.

And who doesn’t want an ultra-quiet PC? Get a cool-running CPU and video card, and maybe, just maybe, your PC can survive on its case fan alone again. With this on a mini-ITX board with an external power supply, a completely fanless, ultra-quick PC might be possible.

And I can see all sorts of applications for this thing for my new employer.

I’m as excited as a puppy when company comes over bearing dog biscuits.

Why my ramdisk techniques don’t work with XP

Dave Farquhar Windows , , , , , , , , , ,

I got a question today in a roundabout way asking about ramdisks in Windows, specifically, where to find my instructions for loading Win98 into a ramdisk, and how to do the same in XP.
I haven’t thought about any of this kind of stuff for more than two years. It seems like two lifetimes.

The original instructions appeared in my book, Optimizing Windows (now in the half-price bin at Amazon.com), and instructions to use DriveSpace to compress the disk appear here. You can get the freeware xmsdisk utility this trick requires from simtel.

These techniques absolutely do not work with Windows NT4, 2000, or XP. Despite the similar name, Windows NT/2000/XP are very different operating systems than Windows 9x. Believe it or not, they’re much more closely related to IBM’s OS/2 than they are to Windows 98. Since there is no DOS laying underneath it all, there’s no easy way to do the trickery that the bootable ramdisk tricks use. What these two tricks do is literally intercept the boot process, copy Windows into the ramdisk, then continue booting.

There’s a $99 piece of software called SuperSpeed that gives the NT-based operating systems this capability. I haven’t used it. I imagine it works using the same principle, hooking into the boot process and moving stuff around before booting continues.

The downside, no matter what OS you use, is the boot time. XP boots in seconds, and my book talks about the trickery necessary to get 95 and 98 to boot in 30 seconds or less. But any time you’re moving a few hundred megs or–yikes–a gig or two of data off a disk into a ramdisk, the boot process is going to end up taking minutes instead.

Is it worth it? For some people, yes. It’s nice to have applications load instantly. A lot of things aren’t CPU intensive. You spend more time waiting for your productivity apps to load than you do waiting for them to do anything. Web browsing and e-mail are generally more bandwidth- and disk-intensive than they are CPU-intensive (although CSS seems determined to change that).

But a lot of games aren’t especially disk-intensive, with the possible exception of when they’re loading a new level. So loading the flavor-of-the-week FPS game into a ramdisk isn’t going to speed it up very much.

Of course, XP is far, far more stable than 98. Windows 9x’s lack of stability absolutely drives me up the wall, and for that matter, I don’t think 2000 or XP are as stable as they should be. Given the choice between XP or 98 in a ramdisk, I’d go for XP, with or without speedup utilities.

I’ve made my choice. As I write, I’m sitting in front of a laptop running 2000 (it’s VPNed into work so I can keep an eye on tape backup jobs) and a desktop PC running Linux. I have a 400 MHz Celeron with Windows 98 on it, but it’s the last Win9x box I have (I think I had 4 at one point when I was writing the aforementioned book). Sometimes I use it to play Baseball Mogul and Railroad Tycoon. Right now it doesn’t even have a keyboard or monitor connected to it.

I guess in a way it feels like hypocrisy, but I wrote the first couple of chapters of that book with a word processor running in Red Hat Linux 5.2 (much to my editor’s chagrin), so I started down that path a long, long time ago.

A semi-easy firewall

Dave Farquhar Linux , , , , , , , ,

A single-floppy firewall mini-distribution can be a quick and easy way to save yourself some money if you’ve got an old PC in a closet not doing anything, assuming you stumble across a combination of hardware that works right.
If you don’t stumble across a combination of hardware that works together, you can just as easily spend a weekend and accomplish nothing but uttering strings of four-letter words in combinations never before heard by mankind.

In case you came here looking for hardware that works, here are a few hints. A 10-megabit PCI NE2000 clone in combination with virtually any 10/100 PCI card ought to work fabulously. A pair of 10/100 PCI cards based on the RealTek 8139 chipset, which includes the majority of today’s inexpensive cards, probably will not. If you’re buying new stuff and want ease of use, get a 3Com card and a cheapie. If you want cheap and a little inconvenience, get a Netgear FA311 or 312 and a Realtek 8139-based card, such as a D-Link DFE-530+ or a Linksys. You’ll have to hunt down and install the natsemi.o module to get the Netgear working; most other inexpensive cards on the market will work with the rtl8139.o driver.

Freesco doesn’t supply a driver for the Intel EtherExpress Pro series out of the box. If you’ve got an EEpro, you can make it work by downloading the module and copying it to the floppy, but don’t rush out to buy one. And yes, the 3Com and Intel chipsets are high-performance chipsets, especially compared to the 8139, but remember, routers are machines that pull packets out of a 1.5-megabit pipe (if you’re lucky) and shove packets down an even smaller pipe. In this application, a $40 big-brand card doesn’t give you any advantage over a no-name card that costs $6 at Newegg.com

While these firewalls will technically work fine even on a 386sx/16, trying to make them work with ISA cards can be a long, difficult road. Used Pentium-75s are dirt cheap (and Pentium-60s and 66s are even cheaper, when you can find them) and they’re a lot less trouble because PCI cards don’t require you to rejumper them or hunt down a plug-and-play configuration disk to find out its IRQ and address. I’ve had the best luck with Pentiums that used an Intel Triton chipset or newer (the 430FX, HX, VX, or TX). I’ve tried a couple of boards that had a SiS chipset of 1995 vintage or so, and I could get one network card or the other working, but not both. I don’t want to generalize and say that based on two isolated incidents that all Taiwanese chipsets are junk for this application–for all I know, the problem could have been the BIOS on those boards–but I’ve done this on a handful of Triton-series boards and done well on all of them, and on two SiS boards and failed. Your mileage will probably vary.

How much memory do you need? 16 megs is sheer luxury.

Once you put all this together, the question becomes whether you use a floppy distribution or a full-blown distribution. If you want peace and quiet and cheap, the answer is pretty easy–use a floppy and pull out whatever hard drive was in there.

A full-out distribution like Red Hat or Debian will give you more versatility. You can run meaningful Web and FTP servers if you want (and your ISP allows it). You can run a caching nameserver to speed up your Web browsing. If you feel adventurous, you can even install the Squid caching proxy and speed up your browsing even more (but either use a SCSI drive or put in a bunch of extra memory and run Squid’s cache out of a ramdisk–Squid’s performance on IDE is, to put it mildly, terrible).

I’m having a hard time finding the documentation on how to set up a second network interface quickly. I believe it involves the file /etc/interfaces and the files /etc/sysconfig/ifconfig.eth0 and .eth1, but I don’t have a Linux box handy to investigate at the moment.

Anyway, I like Debian for this application (of course) because I can easily fit a minimal Debian on a 100-meg hard drive.

Once you get your network cards all working and talking to each other, you can build your firewall using this online tool. I just copy it, then Telnet into my Linux box using PuTTY, fire up a text editor, and right-click in the window to paste.

If you want versatility and quiet and don’t mind spending some cash, pick up a CompactFlash-to-IDE adapter and a CompactFlash card of suitable size. Don’t create a swapfile on the CF card–you’ll quickly burn it up that way. Your system will recognize it as a small IDE drive, giving you silent and reliable solid-state storage on the cheap.

The ultimate DOS boot disk

Dave Farquhar DOS , , , , ,

A little over a year ago, someone issued me a challenge: Make a boot disk containing the Microsoft network client and CD-ROM drivers. The problem is that the network client, plus the DOS boot files, plus a CD-ROM driver and MSCDEX almost always takes up more than 1.44 megs.
So I zipped up as much of the junk as I could and made a boot disk that extracted the Zip file to a ramdisk and connected to the network. I had tons of space left over. So I added some niceties like doskey and a mouse driver. I still had space left over. So then I started hunting down every network driver I could find so that one disk could service the mismash of NICs we’ve bought over the years.

It worked, but adding new drivers was beyond the ability of a lot of my coworkers. And I wanted to add a Windows-style network logon and TCP/IP configuration. I started coding it and some of it worked, but eventually I ran out of time so I abandoned it.

Meanwhile, someone else was doing the same thing, and his results were a lot better.

From the guy who brought you Bart’s Way to Create Bootable CD-ROMs, there’s Bart’s Modular Boot Disk.

To get a disk like mine, all you do is make a bootable floppy on a Windows 9x box, then download Bart’s network packages, including whatever NICs you want to support. Then pop back over to the modboot page and grab all the CD-ROM stuff. I made a disk that supports all of the CD-ROM drives Bart had drivers for, plus a half-dozen or so NICs from 3Com, Intel, and SMC, along with mouse support and doskey. I still had over 100K to spare.

If you find yourself just a little bit short of space, you can use the freeware fdformat to format a disk with just 16 root-directory entries and a large cluster size. Use the commmand fdformat a: /d:16 /c:2. The space that would normally go to the bigger root directory and FAT ends up going to storage capacity instead. But don’t try to run fdformat in Windows–find a Win98 box and boot it in DOS mode.

To make life easier on yourself, you might make the disk, then image a blank and keep the image around for when you want to format a maximum-capacity 1.44-meg disk.

How to pad your resume while meeting chicks.

Dave Farquhar General , , , , , , ,

Padding your resume while meeting chicks. I got a phone call last night offering me just that. Seriously. I didn’t hang up or ask to be taken off the calling list because it was a friend. Not a male friend with a harebrained, sleazy scheme. It was Jeanne. So it was a female friend with a sleazy scheme.
I guess it helps to know Jeanne. She has the distinction of being the only female friend who’s ever offered to lend me a copy of Playboy. She said she bought it for the articles. One of those articles was an interview with some film hunk. Another article was an interview with Aimee Mann. But I think it was all a diabolical plot to see what it would take to get me to read a copy of Playboy in front of her.

This time, Jeanne’s plotting to get me to serve on a committee. She tells me there are virtually no males on the committee. “Sixty to one, Dave! With odds like those you can’t lose!” she said.

Didn’t I hear someone say that about the Red Sox earlier this year?

Let’s change the subject to something more cheerful. How about if I list my qualifications?

1. I’m a male of the species homo sapiens.
2. I’m a sucker for dogs that are smarter than my former landlords my eighth grade science teacher the creeps who dated my sister when I was in college. That’s not every dog I’ve ever seen, but it’s a sizable percentage.

Gatermann says this is the most pathetic thing Jeanne’s ever asked me to do. And yes, Gatermann was there when Jeanne conned me into reading that magazine in front of her. (Yes, I gave in. I had to know what Aimee Mann had to say about Jewel, OK? And yes, her interview was just that–an interview.)

I serve on several committees, few of which work as well as I’d like, so it’s probably a good idea for me to participate, just to see if anyone else knows how to make a committee work right. The time commitment is small, so it just makes sense. In a sick sort of way.

Or maybe you can just say I’m easily finding ways to justify padding my resume while meeting women.

Harry Connick Jr. One of my coworkers pulled out a package he’d just received from Amazon. “I ordered two Harry Connick Jr. CDs,” he said. “This is what they sent.” He whipped out two CDs. They got that much right. But the CDs he received were (drum roll) The Bee Gees and LeAnn Rhimes.

He talked about how much he likes Harry Connick Jr. and how he has two tickets to go see him in some faraway city and he’s bringing a date.

“That’s what you think those tickets are for,” I said. Then, in my best concert-announcer voice, I said, “One night only! The Bee Gees! With very special guest LeAnn Rhimes!”

He glared at me.

Speaking of annoying… I got mail from someone who claims to have invented the “compressed ramdisk” technique I’ve talked about here and in my book, said something at least mildly disparaging about Andre Moreira–one of the other Windows-in-a-ramdisk pioneers–and he says he’s patented the technique, and wants me to download a trial copy of his software and link to it off my site.

I e-mailed him and asked him to set the record straight. It sounded to me like he’s claiming to have invented the compressed ramdisk–something CP/M owners were doing way back in 1984, if not earlier–and he wants free advertising from me for his commercial product.

Now, I could be wrong about that. I was wrong about OS/2 being the next big thing, after all. But if I’ve got the story more or less right, then the answer is no.

Now how did CP/M owners do compressed ramdisks? You’d just put your must-have utilities and applications into an .LBR file, then you’d run SQ on it to compress it. Then in profile.sub–the CP/M equivalent of autoexec.bat–you copied the archive to M: (CP/M’s built-in ramdisk) and then you decompressed it. In the days when applications were smaller than 64K, you could put your OS’ crucial utilities, plus WordStar and dBASE into a ramdisk and smoke all your neighbors who were running that newfangled MS-DOS.

I rediscovered the technique on my Commodore 128 (which was capable of running CP/M) in the late 1980s and thought I was really hot stuff with my 512K ramdisk.

Anyone who thinks the compressed ramdisk was invented in 1999 or 2000 either doesn’t remember his history or is smoking crack.

SCSI! SCSI vs. IDE is a long debate, almost a religious war, and it always has been. I remember seeing SCSI/IDE debates on BBSs in the early 1990s. Few argued that IDE was better than SCSI, though some did–but when you’re using an 8 MHz bus it doesn’t really matter–but IDE generally was less expensive than SCSI. The difference wasn’t always great. I remember seeing an IDE drive sell for $10 less than the SCSI version. The controller might have cost more, but back in the days when a 40-meg drive would set you back $300, a $10 premium for SCSI was nothing. To me, that settled the argument. It didn’t for everyone.

Today, IDE is cheap. Real cheap. A 20-gig drive costs you 50 bucks. A 7200-rpm 40-gig drive is all the drive many people will ever need, and it’s 99 bucks. And for simple computers, that’s great. If it fails, so what? Buy two drives and copy your important data over. At today’s prices you can afford to do that.

SCSI isn’t cheap. It’s hard to find a controller for less than $150, whereas IDE is included free on your motherboard. And if you find a SCSI drive for less than $150, it’s a closeout special. A 20-gig SCSI drive is likely to set you back $175-$200.

Superficially, the difference is philosophy. The IDE drive is designed to be cheap. Good enough to run Word, good enough to play Quake, quiet enough to not wake the baby, cheap enough to sell them by the warehouseful.

SCSI is designed for workstations and servers, where the only things that matter are speed, reliability, speed and speed. (Kind of like spam egg spam and spam in that Monty Python skit). If it costs $1,000 and requires a wind tunnel to cool it and ear protection to use it, who cares? It’s fast! So this is where you see extreme spindle rates like 10,000 and 15,000 RPM and seek times of 4.9 or even 3.9 milliseconds and disk caches of 4, 8, or even 16 MB. It’s also not uncommon to find a 5-year warranty.

In all fairness, I put my Quantum Atlas 10K3 in a Coolermaster cooler. It’s a big bay adapter that acts like a big heatsink and has a single fan, and it also dampens the sound. The setup is no louder than some of the 5400 RPM IDE drives Quantum was manufacturing in 1996-97.

OK, so what’s the practical difference?

IDE is faithful and dumb. You give it requests, it handles them in the order received. SCSI is smart. You send a bunch of read and write requests, and SCSI will figure out the optimal order to execute them in. That’s why you can defrag a SCSI drive while running other things without interrupting the defrag process very much. (Out of order execution is also one of the main things that makes modern CPUs faster than the 486.)

And if you’re running multiple devices, only one IDE device can talk at a time. SCSI devices can talk until you run out of bandwidth. So 160 MB/sec and 320 MB/sec SCSI is actually useful, unlike 133 MB/sec IDE, which is only useful until your drive’s onboard cache empties. Who cares whether a 2-meg cache empties in 0.0303 seconds or 0.01503 seconds?

There’s another advantage to SCSI with multiple devices. With IDE devices, you get two devices per channel, one interrupt per channel. With SCSI, you can do 7 devices per channel and interrupt. Some cards may give you 14. I know a lot of us are awfully crowded for interrupts, so being able to string a ton of devices off a single channel is very appealing. IRQ conflicts are rare these days but they’re not unheard of. SCSI giving you in one interrupt what IDE gives you in four is very nice in a crowded system.

Next up for my dining room… A TV studio.

Dave Farquhar General ,

I did it. I finally did it. I’ve been threatening for a long time to do it. I’ve finally, completely, totally gone off the deep end. And I like it.
I just ordered an 18-gig, 10,000-RPM Maxtor (formerly Quantum) Atlas III hard drive. I ordered an Adaptec Ultra160 host adapter to go with it, since this drive would pretty much saturate my old Adaptec 2940UW. And of course since I’m spending this obnoxious amount of money on a drive–around $200, when a mainstream drive of this size would go for 50 bucks–I’m protecting the investment with a $25 drive cooler. The cooler also deadens the drive’s sound, which is good. I’m a bit nervous about having a 10,000-rpm helicopter in my dining room-turned-office. Hey, where else was I going to put my desk?

I didn’t just get this drive so I could play Civ 3 or compile Linux kernels at blazing speed. I had another reason for this purchase. I also bought a Pinnacle DV500+ video capture/editing card. It’s not a cheap toy, but considering the capabilities it gives you, it’s a steal for the money. I could edit full-length movies with this thing. Well, I could with a capable hard drive. It needs a 25 MB/sec stream to spit out video, and, well, the fastest drive I have won’t do that. Ramdisks? Nice idea, but you can assume a minute of video will chew up a gig of disk, so I’d need 4 GB RAM for most of the projects I have in mind. None of my motherboards will take that much memory.

So my Duron-750 is going to become a video editing workstation. I’ll have to buy or scrounge a bit more memory–Pinnacle recommends 256 MB; I might as well do them a little better and go 384–but then I’ll have the ability to edit video in my dining room. A Duron-750 isn’t much CPU by today’s standards, but Pinnacle lists a P3-500 as the minimum, and the reviews I’ve read do fine with a 500 or 550 MHz CPU. You can assume a Duron runs at a similar speed to a P3 or an Athlon that runs 100 MHz slower, so my Duron-750 should perform like a P3-650. If that proves inadequate, hey, a 1.1 GHz Duron runs $89 these days.

The DV500+ is supposed to be a real bear to set up. We’ll see how it likes my FIC AZ11. I’ve made tricky hardware play before, so I’m not too afraid of this. Every review I’ve read complained about the setup, but once the reviewer got it running, each raved about its abilities.

I can’t wait.

What can I say about Tuesday…?

Dave Farquhar General , , , , , , , , ,

Photography. Tom sent me links to the pictures he took on the roof of Gentry’s Landing a couple of weeks ago. He’s got a shot of downtown, the dome, and the warehouse district, flanked by I-70 on the west and the Mississippi River on the east.
I’m tired. I spent yesterday fighting Mac OS X for a couple of hours. It still feels like beta software. I installed it on a new dual-processor G4/533 with 384 MB RAM, and it took four installation attempts to get one that worked right. Two attempts just flat-out failed, and the installation said so. A third attempt appeared successful, but it felt like Windows 95 on a 16-MHz 386SX with 4 megs of RAM. We’re talking a boot time measured in minutes here. The final attempt was successful and it booted in a reasonable time frame–not as fast as Windows 2000 on similar hardware and nowhere near the 22 seconds I can make Win9x boot in, but faster, I think, than OS 9.1 would boot on the same hardware–and the software ran, but it was sluggish. All the eye candy certainly wasn’t helping. Scrolling around was really fast, but window-resizing was really clunky, and the zooming windows and the menus that literally did drop down from somewhere really got on my nerves.

All told, I’m pretty sure my dual Celeron-500 running Linux would feel faster. Well, I know it’d be faster because I’d put a minimalist GUI on it and I’d run a lot of text apps. But I suspect even if I used a hog of a user interface like Enlightenment, it would still fare reasonably well in comparison.

I will grant that the onscreen display is gorgeous. I’m not talking the eye candy and transparency effects, I’m talking the fonts. They’re all exceptionally crisp, like you’d expect on paper. Windows, even with font smoothing, can’t match it. I haven’t seen Linux with font smoothing. But Linux’s font handling up until recently was hideous.

It’s promising, but definitely not ready for prime time. There are few enough native apps for it that it probably doesn’t matter much anyway.

Admittedly, I had low expectations. About a year ago, someone said something to me about OS X, half in jest, and I muttered back, “If anyone can ruin Unix, it’s Apple.” Well, “ruin” is an awfully harsh word, because it does work, but I suspect a lot of people won’t have the patience to stick with it long enough to get it working, and they may not be willing to take the extreme measures I ultimately took, which was to completely reformat the drive to give it a totally clean slate to work from.

OS X may prove yet to be worth the wait, but anyone who thinks the long wait is over is smoking crack.

Frankly, I don’t know why they didn’t just compile NeXTStep on PowerPC, slap in a Mac OS classic emulation layer, leave the user interface alone (what they have now is an odd hybrid of the NeXT and Mac interfaces that just feels really weird, even to someone like me who’s spent a fair amount of time using both), and release it three years ago.

But there are a lot of things I don’t know.

I spent the rest of the day fighting Linux boot disks. I wanted the Linux equivalent of a DOS boot disk with Ghost on it. Creating one from scratch proved almost impossible for me, so I opted instead to modify an existing one. The disks provided at partimage.org were adequate except they lacked sfdisk for dumping and recreating partition tables. (See Friday if you don’t have the foggiest idea what I’m talking about right about now, funk soul brother.) I dumped the root filesystem to the HD by booting off the two-disk set, mounting the hard drive (mount -t ext2 /dev/hda1 /mnt) and copying each directory (cp -a [directory name] [destination]). Then I made modifications. But nothing would fit, until I discovered the -a switch. The vanilla cp command had been expanding out all the symlinks, bloating the filesystem to a wretched 10 megs. It should have been closer to 4 uncompressed, 1.4 megs compressed. Finally I got what I needed in there and copied it to a ramdisk in preparation for dumping it to a floppy. (You’ve gotta compress it first and make sure it’ll fit.) I think the command was dd if=/dev/ram0 bs=1k | gzip -v9 > [temporary file]. The size was 1.41 MB. Excellent. Dump it to floppy: dd if=[same temporary file from before] of=/dev/fd0 bs=1k

And that’s why my mind feels fried right now. Hours of keeping weird commands like that straight will do it to you. I understand the principles, but the important thing is getting the specifics right.