So I decided I'm gonna fork over the money and go for the 8 core 2.66. After looking at the benchmarks on Macrumors today, I realized the 2.26 sucks... especially when it can't be last year's 2.8 which was the lowest 8 core you could get. And it sucks even harder on single threaded apps. I work off of my macs and make a living off of my computers, so this is something I need. Just so hard to justify that extra 1200 difference between the 2.26 and 2.66. Also the 2.8 from amazon (2600) is out because I need virtualization. All in all I think I'll be very happy with the 2.66 and the 4870. Now to get an IPS panel 24" so I can run dual 24" lcds. Ugh... so much $.
I find it amazing that someone would think a 2.26 GHz machine sucks because it only equals the multithreaded scores of the just now discontinued 3.2 GHz machine, even if the single threaded scores were lower. The second result shows a 2.8. so it's possible that it's in between.
It's clear to see now, in hindsight, that the 1st and 2nd generation Mac Pros were exceptional value for money. In both cases, you were able to purchase a standard configuration which could be upgraded without any false limitations being imposed.
Apple have obviously seen a lot of people purchasing the base configuration and upgrading it with cheaper third party components. As a result, they have reverted to their old ways of crippling the entry level machine and reserving core features for the 'upper' start configuration. I guess Apple want to make their cut up-front.
It reminds me of the stunts that they pulled with the Power Mac G5.
In the next few years, there are going to be a lot of pro users who require 8GB+ of RAM. With the 1st and 2nd generation of Mac Pros, you were looking at £1,749 for a suitable platform. Today you are looking at £2,499. That's a £750 price hike for a feature that should be a given in a professional Mac.
Are there any technical reasons why Apple couldn't have simply left the second processor socket empty and retained the 32GB ceiling in the entry level machine? If not, I fear the good days are over. I fear that 'Sneaky Apple' is back.
If they left the second socket, the machine would have cost even more.
We don't yet know if the machine can hold 16 GB or not.
[QUOTE=emig647;1388166]Personally, I think the 8gb limit should be 16gb from a technical point of view. I think apple intentionally crippled that part. However, I think there is a 16gb limit per chip. Keep in mind the memory controller is built into these cpus now. And I believe the limit of the memory controller is 16gb, though I couldn't find any urls to back that up. However, it does make sense if you think that running dual quad cores and each quad core has 1 memory controller.... 2x16 = 32gb.
I'm thinking that the new quad mac pros have a real expansion of 16gb, but apple isn't selling the 4gb modules with this machine. It wouldn't be the first time apple said the ram limit is xx while the real limit was something else. OWC is famous for finding and selling ram that will go beyond apple's listed "limit".
I really don't see what was so upgradable on the 1st and 2nd gen mac pros. Sure you could upgrade the cpu's, but it cost an arm and a leg. You'd be better off buying a new machine. Not to mention it is a complete pain in the ass to do so. So I suppose you could put more ram in them. Bigger hard drives. The 1st gen gets burned on getting upgraded GPUs, even though NVidia FINALLY came out with a 1st gen 8800gt. Just doesn't seem that upgradable to me. Not like back in the day with the G4's. You could upgrade anything in those. Great machines for great prices then.
With that cinebench image earlier, I'm definitely getting the 2.66 8 core. It just makes sense. Look how poorly that 2.26 does in comparison to machines 2 revs earlier on a single core render.[quote]
It's known that if a program isn't programmed properly, so that its multithreading performance is poor, it won't perform as well on an Nehalem. I'm willing to bet that companies will fix these problems with their programs as soon as they can, so they won't be shown up as running slowly.
As most programs that need the speed are fairly effective at multithreading, they will do better.
I find it amazing that someone would think a 2.26 GHz machine sucks because it only equals the multithreaded scores of the just now discontinued 3.2 GHz machine, even if the single threaded scores were lower. The second result shows a 2.8. so it's possible that it's in between.
I would tend to think it's pretty damn good!
I know you saw the cinebench scores in the url above... the 2.26 was slower in single and slower in multi-threaded than the 2.8 in cinebench. Also the 2.8 beat the 2.26 on a benchmark on macrumors
So that's 2 benchmarks that show the 2.26 being beat by the lowest end 8 core from last year.
So yah, I find it amazing that apple can start off with the 2.26.
Quote:
Originally Posted by melgross
It's known that if a program isn't programmed properly, so that its multithreading performance is poor, it won't perform as well on an Nehalem. I'm willing to bet that companies will fix these problems with their programs as soon as they can, so they won't be shown up as running slowly.
As most programs that need the speed are fairly effective at multithreading, they will do better.
It's easier said than done to optimize a program for multi-threading. Yes it's easy to implement when you design from the ground up, but it becomes more difficult when the app has a 10 year code history behind it. It starts to boil down to a cost / performance issue. If a company can get away without making something fully optimized and still sell the app, they aren't going to focus on making it multi-threaded until they absolutely have to. Also a time factor is involved.
With that being said, I think it will be interesting to see how many NEED to be converted with GCP / OpenCL / Snow leopard around the bend. Only time will tell. In the mean time I'm gonna shoot for the 2.66 since it seems to be the best value for the buck according to what few benchmarks there are to go off of.
I know you saw the cinebench scores in the url above... the 2.26 was slower in single and slower in multi-threaded than the 2.8 in cinebench. Also the 2.8 beat the 2.26 on a benchmark on macrumors
So that's 2 benchmarks that show the 2.26 being beat by the lowest end 8 core from last year.
So yah, I find it amazing that apple can start off with the 2.26.
It's easier said than done to optimize a program for multi-threading. Yes it's easy to implement when you design from the ground up, but it becomes more difficult when the app has a 10 year code history behind it. It starts to boil down to a cost / performance issue. If a company can get away without making something fully optimized and still sell the app, they aren't going to focus on making it multi-threaded until they absolutely have to. Also a time factor is involved.
With that being said, I think it will be interesting to see how many NEED to be converted with GCP / OpenCL / Snow leopard around the bend. Only time will tell. In the mean time I'm gonna shoot for the 2.66 since it seems to be the best value for the buck according to what few benchmarks there are to go off of.
First, to set things straight. I also ordered the dual 2.66 with the 4870. It comes in Friday the 13th. My lucky day in several ways.
Most apps that need the speed, as far as I've seen, already are threaded fairly well, at least. PS will see an improvement there in CS5.
As we both have been saying, apparently, Apple's new tech will make the single threaded apps act better. So that performance will get better.
I think the performance now is pretty good.
These chips, and AMD's new chips as well are optimized for multithreaded performance, and memory bandwidth.
You're not going to find anything different anywhere. It's not Apple, it's the world that's going this way.
But, as you know, these benchmarks aren't representative of the read world. No synthetic benchmarks ever are.
Very true, but I've always felt cinebench is a very good indication of what you can pull from a machine, especially if you're a render farm. If you're constantly compiling, rendering, or encoding... the scores are a great indication of what you should expect. Obviously in real world you have a lot of idle time and human interface time.
I do agree, most of these big time apps are there already. Just makes me nervous to go after a 2.26 when this machine is intended to last 3-4 years. When spending that much $ it better last that long.
Very true, but I've always felt cinebench is a very good indication of what you can pull from a machine, especially if you're a render farm. If you're constantly compiling, rendering, or encoding... the scores are a great indication of what you should expect. Obviously in real world you have a lot of idle time and human interface time.
I do agree, most of these big time apps are there already. Just makes me nervous to go after a 2.26 when this machine is intended to last 3-4 years. When spending that much $ it better last that long.
I've been advising people to go for the 2.66 myself.
But except for one person here who was aghast when I mentioned it, I think it's a good idea to look at the chip offerings Intel will have in late 2010.
While it was a pain, I've replaced dual core cpu's in Mac Pros with the 4 core, faster versions, over a year and a half later. It's what I intend to do with mine when the 6 core 3.2 GHz versions are available.
These machines look much easier to upgrade than the older ones.
Hrm, well it all depends on the cost of the chips. I remember when anandtech upgraded the woodcrest machines to clovertown, it cost a lot more than it was worth. Better off buying the new machine from apple. I mean, you can't really sell chips out of the mac pro that easily. That's a big jump. I'd rather just sell the machine in hand and buy a new one.
I've been waiting for this rev since woodcrest came out. I've waited far too long, so I'm ready to jump on board. I can't wait any longer. It will be a great machine. When do you expect to get yours mel?
Hrm, well it all depends on the cost of the chips. I remember when anandtech upgraded the woodcrest machines to clovertown, it cost a lot more than it was worth. Better off buying the new machine from apple. I mean, you can't really sell chips out of the mac pro that easily. That's a big jump. I'd rather just sell the machine in hand and buy a new one.
I've been waiting for this rev since woodcrest came out. I've waited far too long, so I'm ready to jump on board. I can't wait any longer. It will be a great machine. When do you expect to get yours mel?
Friday the 13th! A lucky day for me over the years
Sweet man. Hope you post your impressions here. At the very least message me. I won't be ordering mine until I get back from St. Louis (business trip). I'm still debating if I should get the apple 24" LCD with it or the HP IPS 24"... I hate glossy so i don't know.
Sweet man. Hope you post your impressions here. At the very least message me. I won't be ordering mine until I get back from St. Louis (business trip). I'm still debating if I should get the apple 24" LCD with it or the HP IPS 24"... I hate glossy so i don't know.
When are you getting back?
I've not bought an Apple monitor since the Studio 21".
I get back the 21st. My ex-business partner bought the aluminum 23" lcd and had nothing but problems with it. It makes me weary buying an apple LCD.
I honestly don't think I've ever bought an apple display (cept the one that came with my performa 630cd).
Not to derail the thread too hard, but this kinda relates to the new mac pro. They come with a DVI and DisplayPort. Has anyone had any issues using the Mini-DisplayPort -> DVI converter? Any loss of color / detail? i heard it goes digital to digital instead of digital to analog to digital in a review on apple's site. Not sure why this would be a bad thing.
I get back the 21st. My ex-business partner bought the aluminum 23" lcd and had nothing but problems with it. It makes me weary buying an apple LCD.
I honestly don't think I've ever bought an apple display (cept the one that came with my performa 630cd).
Not to derail the thread too hard, but this kinda relates to the new mac pro. They come with a DVI and DisplayPort. Has anyone had any issues using the Mini-DisplayPort -> DVI converter? Any loss of color / detail? i heard it goes digital to digital instead of digital to analog to digital in a review on apple's site. Not sure why this would be a bad thing.
What do you recommend for 24" melgross?
The conversion is just minor standards interfacing. The signal isn't touched.
What is your main purpose for the monitor, and what is the budget?
The conversion is just minor standards interfacing. The signal isn't touched.
What is your main purpose for the monitor, and what is the budget?
The main purpose will be my development monitor. I have a 24" samsung TN panel and it blows. Trying to do a web page design sucks as 1 color will be different shades because of the vertical viewing angle. But I mainly do xcode / bbedit coding. I also game a bit, but I can use the samsung for that. (I HAVE to run 2 monitors for the work i do, I'd die without it).
So what are the main points?
Text needs to be crisp
Colors need to be accurate
Viewing angle needs to be non existent (178° would be nice)
Refresh rate needs to be acceptable (no ghosting would be nice).
Since I'll be doing a lot of text, it'd be nice if it wasn't glossy. I can't stand reading on a glossy screen more than a few minutes.
The main purpose will be my development monitor. I have a 24" samsung TN panel and it blows. Trying to do a web page design sucks as 1 color will be different shades because of the vertical viewing angle. But I mainly do xcode / bbedit coding. I also game a bit, but I can use the samsung for that. (I HAVE to run 2 monitors for the work i do, I'd die without it).
So what are the main points?
Text needs to be crisp
Colors need to be accurate
Viewing angle needs to be non existent (178° would be nice)
Refresh rate needs to be acceptable (no ghosting would be nice).
Since I'll be doing a lot of text, it'd be nice if it wasn't glossy. I can't stand reading on a glossy screen more than a few minutes.
As much as I hate to say it, the Dell ultrasharp 2408 WPF is a pretty good monitor, and goes for something in the $600's.
I don't know much about that model Hp.
Well that HP monitor is a H-IPS panel... so that's hard to argue with right away. But just fyi... I'd rather chew glass than use a Dell LCD on my mac pro
I think in the end I'll end up with that HP or the Apple display... probably the HP. What I really want is one of those NEC IPS displays, but I don't want to spend 1200, and they use the same panel as that HP.
Well that HP monitor is a H-IPS panel... so that's hard to argue with right away. But just fyi... I'd rather chew glass than use a Dell LCD on my mac pro
I think in the end I'll end up with that HP or the Apple display... probably the HP. What I really want is one of those NEC IPS displays, but I don't want to spend 1200, and they use the same panel as that HP.
The electronics plays a big part in this, as does some handpicking of the panels.
These new processors may be exciting but they are no match against their price tag!
No surprise why MS took over the market.
When dell and Boxx release their workstations with these chips in them, and they are 3-500 cheaper. Then we'll listen to you. Until then, you don't know the cost involved with these chips. Remember this rev took a new motherboard redesign (new socket / northbridge / ram), completely new chips (that means they start out with low bulk #'s), and apple got them before ANYONE. There is always a cost when being first.
Saw this question over on macrumors and it seemed pertinent here, perhaps someone knows the answer:
Quote:
Looking at the cinebench benchmark, if we divide the single thread scores by the processor speed for the Nehalems we get:
4074/2.93 = 1390 per ghz
3572/2.66 = 1343 per ghz
2039/2.26 = 1022 per ghz
Why is the 2.93 36% faster than would be predicted by the clock difference compared to the 2.26 alone? The 2.26 is even 10% slower than the harpertowns (~1100 per ghz) when compensating for clock frequency. What is going wrong with the 2.26, and shouldn't turbo boost be doing something here?
Sounds like some problem on the 2.26ghz nehalems to me...
Comments
So I decided I'm gonna fork over the money and go for the 8 core 2.66. After looking at the benchmarks on Macrumors today, I realized the 2.26 sucks... especially when it can't be last year's 2.8 which was the lowest 8 core you could get. And it sucks even harder on single threaded apps. I work off of my macs and make a living off of my computers, so this is something I need. Just so hard to justify that extra 1200 difference between the 2.26 and 2.66. Also the 2.8 from amazon (2600) is out because I need virtualization. All in all I think I'll be very happy with the 2.66 and the 4870. Now to get an IPS panel 24" so I can run dual 24" lcds. Ugh... so much $.
I find it amazing that someone would think a 2.26 GHz machine sucks because it only equals the multithreaded scores of the just now discontinued 3.2 GHz machine, even if the single threaded scores were lower. The second result shows a 2.8. so it's possible that it's in between.
I would tend to think it's pretty damn good!
It's clear to see now, in hindsight, that the 1st and 2nd generation Mac Pros were exceptional value for money. In both cases, you were able to purchase a standard configuration which could be upgraded without any false limitations being imposed.
Apple have obviously seen a lot of people purchasing the base configuration and upgrading it with cheaper third party components. As a result, they have reverted to their old ways of crippling the entry level machine and reserving core features for the 'upper' start configuration. I guess Apple want to make their cut up-front.
It reminds me of the stunts that they pulled with the Power Mac G5.
In the next few years, there are going to be a lot of pro users who require 8GB+ of RAM. With the 1st and 2nd generation of Mac Pros, you were looking at £1,749 for a suitable platform. Today you are looking at £2,499. That's a £750 price hike for a feature that should be a given in a professional Mac.
Are there any technical reasons why Apple couldn't have simply left the second processor socket empty and retained the 32GB ceiling in the entry level machine? If not, I fear the good days are over. I fear that 'Sneaky Apple' is back.
If they left the second socket, the machine would have cost even more.
We don't yet know if the machine can hold 16 GB or not.
I'm thinking that the new quad mac pros have a real expansion of 16gb, but apple isn't selling the 4gb modules with this machine. It wouldn't be the first time apple said the ram limit is xx while the real limit was something else. OWC is famous for finding and selling ram that will go beyond apple's listed "limit".
I really don't see what was so upgradable on the 1st and 2nd gen mac pros. Sure you could upgrade the cpu's, but it cost an arm and a leg. You'd be better off buying a new machine. Not to mention it is a complete pain in the ass to do so. So I suppose you could put more ram in them. Bigger hard drives. The 1st gen gets burned on getting upgraded GPUs, even though NVidia FINALLY came out with a 1st gen 8800gt. Just doesn't seem that upgradable to me. Not like back in the day with the G4's. You could upgrade anything in those. Great machines for great prices then.
With that cinebench image earlier, I'm definitely getting the 2.66 8 core. It just makes sense. Look how poorly that 2.26 does in comparison to machines 2 revs earlier on a single core render.[quote]
It's known that if a program isn't programmed properly, so that its multithreading performance is poor, it won't perform as well on an Nehalem. I'm willing to bet that companies will fix these problems with their programs as soon as they can, so they won't be shown up as running slowly.
As most programs that need the speed are fairly effective at multithreading, they will do better.
I find it amazing that someone would think a 2.26 GHz machine sucks because it only equals the multithreaded scores of the just now discontinued 3.2 GHz machine, even if the single threaded scores were lower. The second result shows a 2.8. so it's possible that it's in between.
I would tend to think it's pretty damn good!
I know you saw the cinebench scores in the url above... the 2.26 was slower in single and slower in multi-threaded than the 2.8 in cinebench. Also the 2.8 beat the 2.26 on a benchmark on macrumors
So that's 2 benchmarks that show the 2.26 being beat by the lowest end 8 core from last year.
So yah, I find it amazing that apple can start off with the 2.26.
It's known that if a program isn't programmed properly, so that its multithreading performance is poor, it won't perform as well on an Nehalem. I'm willing to bet that companies will fix these problems with their programs as soon as they can, so they won't be shown up as running slowly.
As most programs that need the speed are fairly effective at multithreading, they will do better.
It's easier said than done to optimize a program for multi-threading. Yes it's easy to implement when you design from the ground up, but it becomes more difficult when the app has a 10 year code history behind it. It starts to boil down to a cost / performance issue. If a company can get away without making something fully optimized and still sell the app, they aren't going to focus on making it multi-threaded until they absolutely have to. Also a time factor is involved.
With that being said, I think it will be interesting to see how many NEED to be converted with GCP / OpenCL / Snow leopard around the bend. Only time will tell. In the mean time I'm gonna shoot for the 2.66 since it seems to be the best value for the buck according to what few benchmarks there are to go off of.
I know you saw the cinebench scores in the url above... the 2.26 was slower in single and slower in multi-threaded than the 2.8 in cinebench. Also the 2.8 beat the 2.26 on a benchmark on macrumors
So that's 2 benchmarks that show the 2.26 being beat by the lowest end 8 core from last year.
So yah, I find it amazing that apple can start off with the 2.26.
It's easier said than done to optimize a program for multi-threading. Yes it's easy to implement when you design from the ground up, but it becomes more difficult when the app has a 10 year code history behind it. It starts to boil down to a cost / performance issue. If a company can get away without making something fully optimized and still sell the app, they aren't going to focus on making it multi-threaded until they absolutely have to. Also a time factor is involved.
With that being said, I think it will be interesting to see how many NEED to be converted with GCP / OpenCL / Snow leopard around the bend. Only time will tell. In the mean time I'm gonna shoot for the 2.66 since it seems to be the best value for the buck according to what few benchmarks there are to go off of.
First, to set things straight. I also ordered the dual 2.66 with the 4870. It comes in Friday the 13th. My lucky day in several ways.
Most apps that need the speed, as far as I've seen, already are threaded fairly well, at least. PS will see an improvement there in CS5.
As we both have been saying, apparently, Apple's new tech will make the single threaded apps act better. So that performance will get better.
I think the performance now is pretty good.
These chips, and AMD's new chips as well are optimized for multithreaded performance, and memory bandwidth.
You're not going to find anything different anywhere. It's not Apple, it's the world that's going this way.
But, as you know, these benchmarks aren't representative of the read world. No synthetic benchmarks ever are.
I do agree, most of these big time apps are there already. Just makes me nervous to go after a 2.26 when this machine is intended to last 3-4 years. When spending that much $ it better last that long.
Very true, but I've always felt cinebench is a very good indication of what you can pull from a machine, especially if you're a render farm. If you're constantly compiling, rendering, or encoding... the scores are a great indication of what you should expect. Obviously in real world you have a lot of idle time and human interface time.
I do agree, most of these big time apps are there already. Just makes me nervous to go after a 2.26 when this machine is intended to last 3-4 years. When spending that much $ it better last that long.
I've been advising people to go for the 2.66 myself.
But except for one person here who was aghast when I mentioned it, I think it's a good idea to look at the chip offerings Intel will have in late 2010.
While it was a pain, I've replaced dual core cpu's in Mac Pros with the 4 core, faster versions, over a year and a half later. It's what I intend to do with mine when the 6 core 3.2 GHz versions are available.
These machines look much easier to upgrade than the older ones.
I've been waiting for this rev since woodcrest came out. I've waited far too long, so I'm ready to jump on board. I can't wait any longer. It will be a great machine. When do you expect to get yours mel?
Hrm, well it all depends on the cost of the chips. I remember when anandtech upgraded the woodcrest machines to clovertown, it cost a lot more than it was worth. Better off buying the new machine from apple. I mean, you can't really sell chips out of the mac pro that easily. That's a big jump. I'd rather just sell the machine in hand and buy a new one.
I've been waiting for this rev since woodcrest came out. I've waited far too long, so I'm ready to jump on board. I can't wait any longer. It will be a great machine. When do you expect to get yours mel?
Friday the 13th! A lucky day for me over the years
Sweet man. Hope you post your impressions here. At the very least message me. I won't be ordering mine until I get back from St. Louis (business trip). I'm still debating if I should get the apple 24" LCD with it or the HP IPS 24"... I hate glossy so i don't know.
When are you getting back?
I've not bought an Apple monitor since the Studio 21".
I honestly don't think I've ever bought an apple display (cept the one that came with my performa 630cd).
Not to derail the thread too hard, but this kinda relates to the new mac pro. They come with a DVI and DisplayPort. Has anyone had any issues using the Mini-DisplayPort -> DVI converter? Any loss of color / detail? i heard it goes digital to digital instead of digital to analog to digital in a review on apple's site. Not sure why this would be a bad thing.
What do you recommend for 24" melgross?
I get back the 21st. My ex-business partner bought the aluminum 23" lcd and had nothing but problems with it. It makes me weary buying an apple LCD.
I honestly don't think I've ever bought an apple display (cept the one that came with my performa 630cd).
Not to derail the thread too hard, but this kinda relates to the new mac pro. They come with a DVI and DisplayPort. Has anyone had any issues using the Mini-DisplayPort -> DVI converter? Any loss of color / detail? i heard it goes digital to digital instead of digital to analog to digital in a review on apple's site. Not sure why this would be a bad thing.
What do you recommend for 24" melgross?
The conversion is just minor standards interfacing. The signal isn't touched.
What is your main purpose for the monitor, and what is the budget?
The conversion is just minor standards interfacing. The signal isn't touched.
What is your main purpose for the monitor, and what is the budget?
The main purpose will be my development monitor. I have a 24" samsung TN panel and it blows. Trying to do a web page design sucks as 1 color will be different shades because of the vertical viewing angle. But I mainly do xcode / bbedit coding. I also game a bit, but I can use the samsung for that. (I HAVE to run 2 monitors for the work i do, I'd die without it).
So what are the main points?
Text needs to be crisp
Colors need to be accurate
Viewing angle needs to be non existent (178° would be nice)
Refresh rate needs to be acceptable (no ghosting would be nice).
Since I'll be doing a lot of text, it'd be nice if it wasn't glossy. I can't stand reading on a glossy screen more than a few minutes.
this is the lcd I was considering:
http://www.newegg.com/Product/Produc...82E16824176104
The main purpose will be my development monitor. I have a 24" samsung TN panel and it blows. Trying to do a web page design sucks as 1 color will be different shades because of the vertical viewing angle. But I mainly do xcode / bbedit coding. I also game a bit, but I can use the samsung for that. (I HAVE to run 2 monitors for the work i do, I'd die without it).
So what are the main points?
Text needs to be crisp
Colors need to be accurate
Viewing angle needs to be non existent (178° would be nice)
Refresh rate needs to be acceptable (no ghosting would be nice).
Since I'll be doing a lot of text, it'd be nice if it wasn't glossy. I can't stand reading on a glossy screen more than a few minutes.
this is the lcd I was considering:
http://www.newegg.com/Product/Produc...82E16824176104
As much as I hate to say it, the Dell ultrasharp 2408 WPF is a pretty good monitor, and goes for something in the $600's.
I don't know much about that model Hp.
As much as I hate to say it, the Dell ultrasharp 2408 WPF is a pretty good monitor, and goes for something in the $600's.
I don't know much about that model Hp.
Well that HP monitor is a H-IPS panel... so that's hard to argue with right away. But just fyi... I'd rather chew glass than use a Dell LCD on my mac pro
I think in the end I'll end up with that HP or the Apple display... probably the HP. What I really want is one of those NEC IPS displays, but I don't want to spend 1200, and they use the same panel as that HP.
Well that HP monitor is a H-IPS panel... so that's hard to argue with right away. But just fyi... I'd rather chew glass than use a Dell LCD on my mac pro
I think in the end I'll end up with that HP or the Apple display... probably the HP. What I really want is one of those NEC IPS displays, but I don't want to spend 1200, and they use the same panel as that HP.
The electronics plays a big part in this, as does some handpicking of the panels.
These new processors may be exciting but they are no match against their price tag!
No surprise why MS took over the market.
When dell and Boxx release their workstations with these chips in them, and they are 3-500 cheaper. Then we'll listen to you. Until then, you don't know the cost involved with these chips. Remember this rev took a new motherboard redesign (new socket / northbridge / ram), completely new chips (that means they start out with low bulk #'s), and apple got them before ANYONE. There is always a cost when being first.
These new processors may be exciting but they are no match against their price tag!
No surprise why MS took over the market.
First, that's not necessarily true about the pricing.
And second is that it has nothing to do with how MS took over the market.
Looking at the cinebench benchmark, if we divide the single thread scores by the processor speed for the Nehalems we get:
4074/2.93 = 1390 per ghz
3572/2.66 = 1343 per ghz
2039/2.26 = 1022 per ghz
Why is the 2.93 36% faster than would be predicted by the clock difference compared to the 2.26 alone? The 2.26 is even 10% slower than the harpertowns (~1100 per ghz) when compensating for clock frequency. What is going wrong with the 2.26, and shouldn't turbo boost be doing something here?
Sounds like some problem on the 2.26ghz nehalems to me...