Why Linux is successful here...
on
Linux and Shrek
·
· Score: 1
The main reason Linux is successful in this industry over similar priced alternatives (ie NT/2k) is the simple fact they can modify it to suit their needs.
It seems most of the studios are writing their own software to add on to the base Maya/Alias/SoftImage/Max package in order to get the exact effect they need. I remember seeing a documentary on what ILM was doing in "The Mummy" that described a lot of the custom stuff written to achieve those effects.
With 'close source' products you don't have the option of getting down into the guts of the system and ripping out code that is just slowing you down (probably there for a normal user or something). You can customize the hell out of your OS to make it sing on the 3D rendering pipe and nothing else!
With cheap hardware (so other Unix vendors can't compete) and full OS customisability you really have a great combo that no one else could come near - after all, I'm sure these guys aren't exactly the types to be calling tech support for anything much.
Linux sounds like it's gonna be huge in the movie industry. Doesn't necessicarily translate anywhere else though - otherwise we'd all be running Irix and SGI on our desktops right now.
"The success of software companies like Red Hat..."
If RHAT is a success, I'd really like to hear their definition of a company in trouble...
Reading their 10k report:
"We have incurred operating losses in six of our previous seven fiscal years, including our most recent fiscal year ended February 28, 2001. We expect to incur significant losses for the foreseeable future, as we substantially increase our sales and marketing, research and development and administrative expenses. In addition, we are investing considerable resources in our Red Hat Network initiative and to expand our professional services offerings. As a result, we cannot be certain when or if we will achieve sustained profitability. Failure to become and remain profitable may adversely affect the market price of our common stock and our ability to raise capital and continue operations."
A company that cannot be certain "when or if we will achieve sustained profitability" is hardly anyone's idea of success! They would have been much better leaving any mention of OSS companies out as they have a rather disturbing tendancy to lose money at the moment - rather focus on the benefits of the OSS coding philosophy and hammer all the IIS hacks home.
Still, I guess they had to say something to make Bob feel good.
That's about what I figured. £400, or even £100 is going to be way above what the market could reasonably expect this device to sell for. My guess is the XBox will be around £150 to £180, and by that time the PS2 will be about £120 so for Nokia to get these machines out, they are either going to make a loss on a sale at £80 each, or lease them out like the current digital TV boxes for £30/month.
(How do you get a £ symbol from a US laptop keyboard? I had to cut/paste)
I never heard the XBox having a 300Mhz CPU. Initial speculation was 600MHz, which wasn't that far off. Remember also that that speculation was over 18 months before launch, not 6 months before launch (as the MT is).
Where did you see it has a LAN port? The specs sheet just says USB and IR, not LAN. You are also assuming that if it does have a LAN port you can get at the data on the HDD, which is not necessarily the case. If the machine is not running any file sharing daemons, and has no shell then you are going to have a very hard time getting your hands on anything.
I still don't see how you are going to be able to buy this as a VCR without Nokia losing money. They are going to have to require a subscription to digital TV with the machine - read their press releases:
i) It is for digital tv, internet access and games - no mention of LAN there.
ii) Games are preinstalled, no mention of user installation.
Figure it out - it's a closed box, not open or useful at all!
It has to be possible to hack as it is using a standard CPU. My guess is they have a minimal bootstrap loader in ROM which looks for a signature on the DVD that you drop in. No signature then it just refuses to run the DVD (aside from as a simple DVD player).
It is also possible that the DVDs will be encrypted (my guess is using some standard 128 bit crypto - MS already has the code to do this in Windows with CryptoAPI). Remember MS is a software company and isn't going to use the same crappy crypto as CSS - look for something equivalent to AES or TwoFish).
All of the above could be defeated through the simple act of replacing the boot ROMs. I'd be assuming they will use standard surface mount flash ROMs, so all you need to do is unsolder a few surface mount chips, solder a few more on and you have an XBox running anything you want.
Of course, desoldering surface mount BGA chips is probably gonna be tough...
Look at the specs - it has crappy 3D, something basically equivalent to a TNT or Voodoo2.
I'd be expecting to never see one of these for sale on it's own - why? Because this is Nokia we are talking about and when was the last time you saw a cellphone for sale without a network contract.
These boxes will probably be given away for free as set-top boxes to replace your current digital cable box. You'll probably only get them with a $50/month subscription to a high end cable plan, and then it's quite likely that all content (including games) will come down the cable and not be bought on the shelves, or even be available to run.
Sure, a few geeks will add their own CDROM drives, figure out how to get external titles in and so on, but that will be the vast minority, and remember, after 12 months this low end TV decoder will have cost MORE than that XBox with a few orders of magnitude better graphics processing.
I'm not optimistic about the whole games thing on the Media Console.
Look at the specs, people. This is a machine that is comparable (only just) to the gaming PCs of about 2 years ago!! It has a Celeron 366, a TNT/Voodoo2 equivalent 3D card, a maximum resolution of PAL (720x576) standard, which most Americans can't use anyway and have to deal with NTSC (640x480) and a maximum frame rate of 30fps with some rather strict rules on color usage. There isn't even anti-aliasing to smooth out the bleeding on the TV.
Fact is you will be NOT be getting the latest games, more likely remakes of Gameboy stuff or Quake 1 re-releases and more importantly you will NOT be making a dent in XBox, Gamecube or PS2 sales.
Nokia will not be able to make any money from the direct sales of these boxes either (they would have to sell for more than the vastly superior XBox or PS/2 for this to happen), so expect to get them leased to you for about $50/month or more (along with a digital TV connection). Nokia is simply USING the open source community as a group of people who can provide gimmicks to sell their digital TV boxes.
So, don't expect to EVER see these on shelves but expect some cable guy to drop one on your desk and the games will be much the same distraction you get on the current Nokia phones - a gimmick and a long way from the real purpose of the box in the first place, which is to pipe an encrypted stream of media to your TV, decoded by a proprietary smart-card all for the monthly hit to the cheque book.
To sum up, this box WILL end up costing a substantial subscription fee, probably will never belong to you and the instant you stop paying they will come and take your slow TV-game box away. It isn't a games machine, it will never be a games machine and is going to do nothing to MS/AOL/Sony/Nintendo's bottom line.
1. I checked out Compaq's site and there is about a $10,000 difference between a 4 way system with the Profusion chipset and without it. That's about the cost of 2 CPUs, so you aren't exactly correct here but my initial point wasn't exactly right either.
2. True enough. The Linux machine also included the cost of the UPS which wasn't part of the Compaq system.
3. Interestingly enough, the cost of software on the Linux system (including maintenance) was actually $365,000 where the Windows system was only $27,000... So much for MS being expensive!!
I agree with you on TCO - the initial cost of commercial software is negligible when you factor in usability over the life of the system, unless you are buying software worth many thousands of dollars per user. Most people ignore the cost of administration and downtime when it comes to the purchase of a system - that is usually well over 90% of the cost!!
Note that I've avoided saying which is better. Microsoft's claims are that Windows requires less "real" admin effort than Unix, hence their system is cheaper regardless of the initial cost. *IF* their initial assertion is correct then they are right. If Win2k requires the same amount of admin then they are wrong. IMHO they tend to rely on a premise that once the system is up there is less need to call on a "real" admin for mundane tasks than Unix. Personally I feel they are correct as they have put a lot of work into their admin wizards etc., but YMMV.
As for Linux being in the top 10 in these results, remember that there *are* only 10 results so it's not much of an achievement. The Linux result is the worst price/performance result on record - at least half as expensive again. Also remember TPC scores include maintenance and admin costs in their reports, so the prices aren't that far from real-world, even if the scores themselves are somewhat aggressively inflated.
I'm glad there are people recommending Linux to those in grey suits. Like I said, I tend to prefer Windows but I've been known to toss in a Linux machine here and there when paying the licence fees for an extra box just didn't make sense. The more I work in the industry, the more I believe that getting something working is more important than getting it perfect first time.
This is basically a Celeron 366 with a TNT(1) or Voodoo 2 equivalent - standard PC of about 2 years ago. In fact, Nokia could probably pick these up at a second hand shop for less than the money they will sell them for!!
My guess would be that it is using a TNT2m64 on board (they are very cheap), a very cheap RAMDAC (230MHz according to spec) and a CPU that Intel don't even make any more (slowest Celeron for sale is 667MHz @ $69). Probably going to sell in the $100 price range as it will have to seriously undercut the PS2 and XBox to make any sales at all.
I guess you'd get ok frame rates from Quake 2 or earlier, but don't expect any of the latest games to EVER come out on this box - it will be strictly limited to GameBoy style things.
So basically you are saying that 4 linux machines can beat one Win2k machine. In fact, Linux is close to half the price/performance of any of the machines there.
What it comes down to is that if you want to save money then Win2k is going to give you the best performance for a given price, up to 8CPUs. As there are no benchmarks for a Win2k cluster in TPC-H, you can't draw any conclusions from this 'win'.
If you look in TPC-C however, you'll see that a Win2kAS clusters are ramming Unix right up the hole in performance and price... I'd be sticking to the 'Do TPC benchmarks really mean anything' stories if you want to promote Unix over Windows.
1. You are comparing 4x4CPU boxes with 1x8CPU box. In general, the 8 way machine will cost a lot more than double a 4 way machine - especially on Intel hardware where the CPUs support only 4 way SMP natively. This means the Win2k solution is actually a lot more cost effective that it appears.
2. Just check the disclosure reports - they both use RAID arrays. The Win2k solution was hot-plug, looks like the Linux one wasn't. Oops, score another for Win2k.
3. Right on the money. OS cost compared to hardware cost and support cost is negligible (something MS has been saying for a long time with the magic acronym 'TCO'). In this case though, it looks like Linux still has a way to catch up with Win2k on the same hardware, or even achieve the same price/performance ratio. It is good to see it make it on the scores at last though.
Good point. Taking a long look at the specs, this seems to basically be a PC type device with a TV as the output device (and all the associated resolution problems), and a highly customised window and input manager.
I seriously doubt it will in any way be a competitor to the XBox or PS2 from the way the specs are reading - the main problem being the lack of a firm hardware specification. It's fine for the applications to have a nice solid API foundation, but console games really require direct access to a well defined set of hardware to work well.
If this is billed as a game console, it will fail. It seems much more like a digital TV box than anything else I can imagine. In that market, it might well work but the end user will probably notice it is running Linux about as much as you notice the OS that the fuel injection system on your car is running. In fact, the only time you would see the OS is either in a splash screen or a kernel panic.
I appreciate that Nokia has the resources to make this machine (ie fund the dev costs), I just don't figure that they can make a competitive machine to the PS2 or XBox for less than the selling price. If there was indeed a way to do this then I'm sure that Sony, Nintendo or Microsoft would have found it and rolled it out by now.
If they are trying to offset something on the annual report, then I wouldn't bet on this being a successful product at all. It would probably just make it to prototype and then they'll write the whole thing off when the figures don't add up. In that case it is actually bad for Linux as it ends up another high profile failed product/business that is somehow related to Linux or Open Software.
Open Source does not automatically mean no profit, it just means you have to generate profit from something other than the software or SDK. This is currently not how the game industry works and the only other source of revenue would be the sale of the hardware itself, which I doubt could be profitiable in an arena where the average selling price is 60 to 70% of the manufacture cost.
You actually aren't being so theoretical about saving dev costs from "Off the shelf" and Open Source - that's almost the XBox theory (except trade Open Source with 'we already have an OS we can use').
The problem here is almost so obvious that it is being overlooked by most people: this can't work because of the pricing mechanics of the game industry.
In order to produce a product at a competitive price you have to LOSE money on the sale of the hardware to make it up again on either the license fees from software sold or from the SDK. It has been estimated that it will cost Microsoft about $425 for each XBox but they have to sell them at $300 or less for it to sell against the PS2.
Now if Nokia has an open design then no one will clone it because they would have to lose money from the sale of their hardware with no way to get the money back from any sort of licensing, and by using a GPL operating system, Nokia has no practical way of recovering their loss on the sale of the systems. The best they could do is have a binary-only kernel module which they charge software developers to use, which breaks the whole idea of using a GPL system in the first place.
So, by producing an open hardware console with an open hardware system, Nokia (by my reckoning at least - feel free to correct me) are going to be heading down the long road to a failed project. There is simply no way for them to make money, even if they capture the entire market they are only guaranteed to lose money!!
Somebody told me I shouldn't log in as root all the time, so I just changed my.bashrc to have a 'su -' at the end instead, and then set root's password to nothing.
They way I see it, the Linux community will probably follow the lead of Red Hat (after all they are one of the most influential distos around). This means the support of Netscape itself in the Linux world will severely decline. Soon you'll have different browsers in the Linux world to those found in the corporate mainstream.
Seems to me that this is a very good recipie to get the whole Mozilla/Netscape project to lose all funding completely and go down the tubes. I guess I'll start looking to KDE to provide the browser of the future now and ditch any ideas I had of a resurgence of Netscape on any system anytime soon.
Once again, this is basically admitting the failure of the open source community to successfully cross the bridge into the corporate world - in fact it's more like they promised to cross the bridge, waited for the corporate types to be on the Netscape side and then burned the bridge.
Tuning something to run quickly is a lot easier than writing software to heave shitloads of data for years without ever dropping a single bit.
What complete drivel. Have you ever written any optimized code,
compared to writing sane easy to debug code? I'll tell you from a decade
of experience that it is far easier to write dog-slow safe code than it is to
write well optimized code. Of course, you should always write optimized
code to not lose data as well, but you seem to be inferring that the two are
mutually exclusive (which is also incorrect)
I used the TPC scores to show that MSSQL is in a similar league to Oracle in
the mid-size arena. As for "dropping a single bit", I've not
seen MSSQL drop any more bits than Oracle so you can take your rhetoric and
stick it.
I'm not sure what you are basing the comment that the "outlining architecture is pretty much the same".
Basically, the storage engine is different (no more devices, just files that can autogrow), the query parser and optimiser is new, the stored proc engine is different, indexes were completely uprooted and changed (look at the escalation and row level locking policy changes between 6.5 and 7.0), the interface between the SQL engine and the storage engine was published and made modular...
Actually, I find it hard to think of anything that DIDN'T change going from 6.5 to 7.0. Even most (serious) T-SQL scripts had to change!!
MSSQL Server was Sybase back in the 4.2 incarnation. Since then the product has been totally rewritten (in fact I think there was a little celebration at MS a while back when the last line of Sybase code was removed).
MSSQL is hardly a 'desktop' system either - check out some of the TPC scores it has been getting lately. They've really done a lot of good work in the storage engine recently. I don't believe the Intel hardware will scale to the same level as Sun and other hardware will under Oracle, but from my experience in the database space up to 10G to 100G and less than a few thousand users, MSSQL runs very well.
Oh, I should also mention that MSSQL is a *lot* more developer friendly than Oracle. It really is a nice product and those who haven't used it should have a look before the knee-jerk MS == evil/slow/whatever kicks in.
Microsoft starts auditing to enforce it's licenses and Slashdot runs around expounding on how evil they are and how we must all turn to the light to stop the coming of Satan.
The FSF starts a GPL crackdown and the person that broke the license is the bad guy, not the FSF.
Perhaps you people need to know that people in glass houses shouldn't throw stones?
The Australian law isn't quite as bad as the DMCA. For example:
While making and importing decoding devices will be banned, their personal use will not.
In other words, if you can get your hands on a device, you are allowed to use it
The full text of the modifications to the copyright bill can be found here and contain some interesting definitions which make the whole DeCSS thing quite different in Australia:
circumvention device means a device (including a computer program) having only a limited commercially significant purpose or use, or no such purpose or use, other than the circumvention, or facilitating the circumvention, of an effective technological protection measure
Under that definition, the DeCSS code itself is definitely illegal, but any Linux DVD player is fine because it definitely has a purpose other than simple decryption of the video data. Basically while this bill is still an impediment to the free trade of ideas, I think there is enough restraint in it to make some worthwhile things come out. Hopefully this means that while decss.c might be outlawed, css-auth.c is quite ok.
Slashdot Mirror
The main reason Linux is successful in this industry over similar priced alternatives (ie NT/2k) is the simple fact they can modify it to suit their needs.
It seems most of the studios are writing their own software to add on to the base Maya/Alias/SoftImage/Max package in order to get the exact effect they need. I remember seeing a documentary on what ILM was doing in "The Mummy" that described a lot of the custom stuff written to achieve those effects.
With 'close source' products you don't have the option of getting down into the guts of the system and ripping out code that is just slowing you down (probably there for a normal user or something). You can customize the hell out of your OS to make it sing on the 3D rendering pipe and nothing else!
With cheap hardware (so other Unix vendors can't compete) and full OS customisability you really have a great combo that no one else could come near - after all, I'm sure these guys aren't exactly the types to be calling tech support for anything much.
Linux sounds like it's gonna be huge in the movie industry. Doesn't necessicarily translate anywhere else though - otherwise we'd all be running Irix and SGI on our desktops right now.
"The success of software companies like Red Hat..."
If RHAT is a success, I'd really like to hear their definition of a company in trouble...
Reading their 10k report:
"We have incurred operating losses in six of our previous seven fiscal years, including our most recent fiscal year ended February 28, 2001. We expect to incur significant losses for the foreseeable future, as we substantially increase our sales and marketing, research and development and administrative expenses. In addition, we are investing considerable resources in our Red Hat Network initiative and to expand our professional services offerings. As a result, we cannot be certain when or if we will achieve sustained profitability. Failure to become and remain profitable may adversely affect the market price of our common stock and our ability to raise capital and continue operations."
A company that cannot be certain "when or if we will achieve sustained profitability" is hardly anyone's idea of success! They would have been much better leaving any mention of OSS companies out as they have a rather disturbing tendancy to lose money at the moment - rather focus on the benefits of the OSS coding philosophy and hammer all the IIS hacks home.
Still, I guess they had to say something to make Bob feel good.
That's about what I figured. £400, or even £100 is going to be way above what the market could reasonably expect this device to sell for. My guess is the XBox will be around £150 to £180, and by that time the PS2 will be about £120 so for Nokia to get these machines out, they are either going to make a loss on a sale at £80 each, or lease them out like the current digital TV boxes for £30/month.
(How do you get a £ symbol from a US laptop keyboard? I had to cut/paste)
I never heard the XBox having a 300Mhz CPU. Initial speculation was 600MHz, which wasn't that far off. Remember also that that speculation was over 18 months before launch, not 6 months before launch (as the MT is).
Where did you see it has a LAN port? The specs sheet just says USB and IR, not LAN. You are also assuming that if it does have a LAN port you can get at the data on the HDD, which is not necessarily the case. If the machine is not running any file sharing daemons, and has no shell then you are going to have a very hard time getting your hands on anything.
I still don't see how you are going to be able to buy this as a VCR without Nokia losing money. They are going to have to require a subscription to digital TV with the machine - read their press releases:
i) It is for digital tv, internet access and games - no mention of LAN there.
ii) Games are preinstalled, no mention of user installation.
Figure it out - it's a closed box, not open or useful at all!
It has to be possible to hack as it is using a standard CPU. My guess is they have a minimal bootstrap loader in ROM which looks for a signature on the DVD that you drop in. No signature then it just refuses to run the DVD (aside from as a simple DVD player).
It is also possible that the DVDs will be encrypted (my guess is using some standard 128 bit crypto - MS already has the code to do this in Windows with CryptoAPI). Remember MS is a software company and isn't going to use the same crappy crypto as CSS - look for something equivalent to AES or TwoFish).
All of the above could be defeated through the simple act of replacing the boot ROMs. I'd be assuming they will use standard surface mount flash ROMs, so all you need to do is unsolder a few surface mount chips, solder a few more on and you have an XBox running anything you want.
Of course, desoldering surface mount BGA chips is probably gonna be tough...
Look at the specs - it has crappy 3D, something basically equivalent to a TNT or Voodoo2.
I'd be expecting to never see one of these for sale on it's own - why? Because this is Nokia we are talking about and when was the last time you saw a cellphone for sale without a network contract.
These boxes will probably be given away for free as set-top boxes to replace your current digital cable box. You'll probably only get them with a $50/month subscription to a high end cable plan, and then it's quite likely that all content (including games) will come down the cable and not be bought on the shelves, or even be available to run.
Sure, a few geeks will add their own CDROM drives, figure out how to get external titles in and so on, but that will be the vast minority, and remember, after 12 months this low end TV decoder will have cost MORE than that XBox with a few orders of magnitude better graphics processing.
I'm not optimistic about the whole games thing on the Media Console.
Look at the specs, people. This is a machine that is comparable (only just) to the gaming PCs of about 2 years ago!! It has a Celeron 366, a TNT/Voodoo2 equivalent 3D card, a maximum resolution of PAL (720x576) standard, which most Americans can't use anyway and have to deal with NTSC (640x480) and a maximum frame rate of 30fps with some rather strict rules on color usage. There isn't even anti-aliasing to smooth out the bleeding on the TV.
Fact is you will be NOT be getting the latest games, more likely remakes of Gameboy stuff or Quake 1 re-releases and more importantly you will NOT be making a dent in XBox, Gamecube or PS2 sales.
Nokia will not be able to make any money from the direct sales of these boxes either (they would have to sell for more than the vastly superior XBox or PS/2 for this to happen), so expect to get them leased to you for about $50/month or more (along with a digital TV connection). Nokia is simply USING the open source community as a group of people who can provide gimmicks to sell their digital TV boxes.
So, don't expect to EVER see these on shelves but expect some cable guy to drop one on your desk and the games will be much the same distraction you get on the current Nokia phones - a gimmick and a long way from the real purpose of the box in the first place, which is to pipe an encrypted stream of media to your TV, decoded by a proprietary smart-card all for the monthly hit to the cheque book.
To sum up, this box WILL end up costing a substantial subscription fee, probably will never belong to you and the instant you stop paying they will come and take your slow TV-game box away. It isn't a games machine, it will never be a games machine and is going to do nothing to MS/AOL/Sony/Nintendo's bottom line.
If you compare the Linux solution to the next Win2k solution, check out the software prices:
Linux + DB2 = $366,000
Win2k + SQL2000 = $27,000
OMG!!! No wonder Microsoft wins sales in the database arena!
1. I checked out Compaq's site and there is about a $10,000 difference between a 4 way system with the Profusion chipset and without it. That's about the cost of 2 CPUs, so you aren't exactly correct here but my initial point wasn't exactly right either.
2. True enough. The Linux machine also included the cost of the UPS which wasn't part of the Compaq system.
3. Interestingly enough, the cost of software on the Linux system (including maintenance) was actually $365,000 where the Windows system was only $27,000... So much for MS being expensive!!
I agree with you on TCO - the initial cost of commercial software is negligible when you factor in usability over the life of the system, unless you are buying software worth many thousands of dollars per user. Most people ignore the cost of administration and downtime when it comes to the purchase of a system - that is usually well over 90% of the cost!!
Note that I've avoided saying which is better. Microsoft's claims are that Windows requires less "real" admin effort than Unix, hence their system is cheaper regardless of the initial cost. *IF* their initial assertion is correct then they are right. If Win2k requires the same amount of admin then they are wrong. IMHO they tend to rely on a premise that once the system is up there is less need to call on a "real" admin for mundane tasks than Unix. Personally I feel they are correct as they have put a lot of work into their admin wizards etc., but YMMV.
As for Linux being in the top 10 in these results, remember that there *are* only 10 results so it's not much of an achievement. The Linux result is the worst price/performance result on record - at least half as expensive again. Also remember TPC scores include maintenance and admin costs in their reports, so the prices aren't that far from real-world, even if the scores themselves are somewhat aggressively inflated.
I'm glad there are people recommending Linux to those in grey suits. Like I said, I tend to prefer Windows but I've been known to toss in a Linux machine here and there when paying the licence fees for an extra box just didn't make sense. The more I work in the industry, the more I believe that getting something working is more important than getting it perfect first time.
This is basically a Celeron 366 with a TNT(1) or Voodoo 2 equivalent - standard PC of about 2 years ago. In fact, Nokia could probably pick these up at a second hand shop for less than the money they will sell them for!!
My guess would be that it is using a TNT2m64 on board (they are very cheap), a very cheap RAMDAC (230MHz according to spec) and a CPU that Intel don't even make any more (slowest Celeron for sale is 667MHz @ $69). Probably going to sell in the $100 price range as it will have to seriously undercut the PS2 and XBox to make any sales at all.
I guess you'd get ok frame rates from Quake 2 or earlier, but don't expect any of the latest games to EVER come out on this box - it will be strictly limited to GameBoy style things.
So basically you are saying that 4 linux machines can beat one Win2k machine. In fact, Linux is close to half the price/performance of any of the machines there.
What it comes down to is that if you want to save money then Win2k is going to give you the best performance for a given price, up to 8CPUs. As there are no benchmarks for a Win2k cluster in TPC-H, you can't draw any conclusions from this 'win'.
If you look in TPC-C however, you'll see that a Win2kAS clusters are ramming Unix right up the hole in performance and price... I'd be sticking to the 'Do TPC benchmarks really mean anything' stories if you want to promote Unix over Windows.
In response,
1. You are comparing 4x4CPU boxes with 1x8CPU box. In general, the 8 way machine will cost a lot more than double a 4 way machine - especially on Intel hardware where the CPUs support only 4 way SMP natively. This means the Win2k solution is actually a lot more cost effective that it appears.
2. Just check the disclosure reports - they both use RAID arrays. The Win2k solution was hot-plug, looks like the Linux one wasn't. Oops, score another for Win2k.
3. Right on the money. OS cost compared to hardware cost and support cost is negligible (something MS has been saying for a long time with the magic acronym 'TCO'). In this case though, it looks like Linux still has a way to catch up with Win2k on the same hardware, or even achieve the same price/performance ratio. It is good to see it make it on the scores at last though.
Good point. Taking a long look at the specs, this seems to basically be a PC type device with a TV as the output device (and all the associated resolution problems), and a highly customised window and input manager.
I seriously doubt it will in any way be a competitor to the XBox or PS2 from the way the specs are reading - the main problem being the lack of a firm hardware specification. It's fine for the applications to have a nice solid API foundation, but console games really require direct access to a well defined set of hardware to work well.
If this is billed as a game console, it will fail. It seems much more like a digital TV box than anything else I can imagine. In that market, it might well work but the end user will probably notice it is running Linux about as much as you notice the OS that the fuel injection system on your car is running. In fact, the only time you would see the OS is either in a splash screen or a kernel panic.
I appreciate that Nokia has the resources to make this machine (ie fund the dev costs), I just don't figure that they can make a competitive machine to the PS2 or XBox for less than the selling price. If there was indeed a way to do this then I'm sure that Sony, Nintendo or Microsoft would have found it and rolled it out by now.
If they are trying to offset something on the annual report, then I wouldn't bet on this being a successful product at all. It would probably just make it to prototype and then they'll write the whole thing off when the figures don't add up. In that case it is actually bad for Linux as it ends up another high profile failed product/business that is somehow related to Linux or Open Software.
Open Source does not automatically mean no profit, it just means you have to generate profit from something other than the software or SDK. This is currently not how the game industry works and the only other source of revenue would be the sale of the hardware itself, which I doubt could be profitiable in an arena where the average selling price is 60 to 70% of the manufacture cost.
You actually aren't being so theoretical about saving dev costs from "Off the shelf" and Open Source - that's almost the XBox theory (except trade Open Source with 'we already have an OS we can use').
The problem here is almost so obvious that it is being overlooked by most people: this can't work because of the pricing mechanics of the game industry.
In order to produce a product at a competitive price you have to LOSE money on the sale of the hardware to make it up again on either the license fees from software sold or from the SDK. It has been estimated that it will cost Microsoft about $425 for each XBox but they have to sell them at $300 or less for it to sell against the PS2.
Now if Nokia has an open design then no one will clone it because they would have to lose money from the sale of their hardware with no way to get the money back from any sort of licensing, and by using a GPL operating system, Nokia has no practical way of recovering their loss on the sale of the systems. The best they could do is have a binary-only kernel module which they charge software developers to use, which breaks the whole idea of using a GPL system in the first place.
So, by producing an open hardware console with an open hardware system, Nokia (by my reckoning at least - feel free to correct me) are going to be heading down the long road to a failed project. There is simply no way for them to make money, even if they capture the entire market they are only guaranteed to lose money!!
:)
Thanks. I probably deserved that.
Sure!! 127.0.0.1 is the one I use but I'm told that if you have one of those new-fangled name resolution things going I can use 'localhost' as well.
;)
Somebody told me I shouldn't log in as root all the time, so I just changed my .bashrc to have a 'su -' at the end instead, and then set root's password to nothing.
Is that bad?
They way I see it, the Linux community will probably follow the lead of Red Hat (after all they are one of the most influential distos around). This means the support of Netscape itself in the Linux world will severely decline. Soon you'll have different browsers in the Linux world to those found in the corporate mainstream.
Seems to me that this is a very good recipie to get the whole Mozilla/Netscape project to lose all funding completely and go down the tubes. I guess I'll start looking to KDE to provide the browser of the future now and ditch any ideas I had of a resurgence of Netscape on any system anytime soon.
Once again, this is basically admitting the failure of the open source community to successfully cross the bridge into the corporate world - in fact it's more like they promised to cross the bridge, waited for the corporate types to be on the Netscape side and then burned the bridge.
To sum it up... ACK!!!
Tuning something to run quickly is a lot easier than writing software to heave shitloads of data for years without ever dropping a single bit.
What complete drivel. Have you ever written any optimized code, compared to writing sane easy to debug code? I'll tell you from a decade of experience that it is far easier to write dog-slow safe code than it is to write well optimized code. Of course, you should always write optimized code to not lose data as well, but you seem to be inferring that the two are mutually exclusive (which is also incorrect)
I used the TPC scores to show that MSSQL is in a similar league to Oracle in the mid-size arena. As for "dropping a single bit", I've not seen MSSQL drop any more bits than Oracle so you can take your rhetoric and stick it.
I'm not sure what you are basing the comment that the "outlining architecture is pretty much the same".
Basically, the storage engine is different (no more devices, just files that can autogrow), the query parser and optimiser is new, the stored proc engine is different, indexes were completely uprooted and changed (look at the escalation and row level locking policy changes between 6.5 and 7.0), the interface between the SQL engine and the storage engine was published and made modular...
Actually, I find it hard to think of anything that DIDN'T change going from 6.5 to 7.0. Even most (serious) T-SQL scripts had to change!!
MSSQL Server was Sybase back in the 4.2 incarnation. Since then the product has been totally rewritten (in fact I think there was a little celebration at MS a while back when the last line of Sybase code was removed).
MSSQL is hardly a 'desktop' system either - check out some of the TPC scores it has been getting lately. They've really done a lot of good work in the storage engine recently. I don't believe the Intel hardware will scale to the same level as Sun and other hardware will under Oracle, but from my experience in the database space up to 10G to 100G and less than a few thousand users, MSSQL runs very well.
Oh, I should also mention that MSSQL is a *lot* more developer friendly than Oracle. It really is a nice product and those who haven't used it should have a look before the knee-jerk MS == evil/slow/whatever kicks in.
Microsoft starts auditing to enforce it's licenses and Slashdot runs around expounding on how evil they are and how we must all turn to the light to stop the coming of Satan.
The FSF starts a GPL crackdown and the person that broke the license is the bad guy, not the FSF.
Perhaps you people need to know that people in glass houses shouldn't throw stones?
The Australian law isn't quite as bad as the DMCA. For example:
While making and importing decoding devices will be banned, their personal use will not.
In other words, if you can get your hands on a device, you are allowed to use it
The full text of the modifications to the copyright bill can be found here and contain some interesting definitions which make the whole DeCSS thing quite different in Australia:
circumvention device means a device (including a computer program) having only a limited commercially significant purpose or use, or no such purpose or use, other than the circumvention, or facilitating the circumvention, of an effective technological protection measure
Under that definition, the DeCSS code itself is definitely illegal, but any Linux DVD player is fine because it definitely has a purpose other than simple decryption of the video data. Basically while this bill is still an impediment to the free trade of ideas, I think there is enough restraint in it to make some worthwhile things come out. Hopefully this means that while decss.c might be outlawed, css-auth.c is quite ok.