Looking to the future...

[roguerower]

I'm looking to build a new PC at the end of the summer, around the beginning of August. Now my problem is that I'm unsure about whether or not to go ahead with it or wait for Nehalem.

I understand that just b/c Nehalem comes out Core isn't going to disappear overnight, but I'd like to stay on top of the technology so I don't get screwed over.

My current plan was to build the system and then wait a year or so until the prices settled, then rip out the Mobo, CPU and RAM (possibly...heard it was going to DDR3) and then sell them and use the money to go with a new set.

Can anyone share their $0.02 on this and maybe tell me whether it's a stupid idea?

 

[rip]

There's NO planning ahead. Decide in August.

 

[core2slow]

thread like these never gets old.

Build something you like NOW.

when nehalem comes out, build something you like THEN.

 

[bryanW1995]

It's fun watching the n00bs rip on the old hand...

Normally I'd say build a ddr3 c2d rig in august, but ddr2 is so unbelievably cheap that I wouldn't bother with it. ddr3 will drop WAY down when nehalem comes out b/c they won't even offer that in ddr2 iirc PLUS amd should have am3 out around the same time.

Having said that, I'd snag a Q9450, a cheap p35 rig (either ip35e or ds3L if they're still around, maybe go p45 if p35 is gone by then) and 2x1gb ddr2. OC the hell out of it but keep voltages reasonable (nothing over 1.4vcore) and you'll get 3.4-3.6ghz. This will clobber the everliving crap out of most nehalems for a year or two, anyway, allowing the early-adopters to sort through all the bugs in nehalem v1.0.

 

[Foxery]

There's always something new on the way. Buy what's available when you want it.

Also, Nehalem Servers are planned to come out before Desktops. It may be a longer wait than you realize.

 

[roguerower]

Jeez, the first two ripped a bit hard, I understand there's no planning, however if there's anyone with any insight, like foxery, I'd like to know that.

I know to build now instead of waiting but there's no shame in asking other people for their opinions.

 

[Idontcare]

Originally posted by: bryanW1995
It's fun watching the n00bs rip on the old hand...

Normally I'd say build a ddr3 c2d rig in august, but ddr2 is so unbelievably cheap that I wouldn't bother with it. ddr3 will drop WAY down when nehalem comes out b/c they won't even offer that in ddr2 iirc PLUS amd should have am3 out around the same time.

Having said that, I'd snag a Q9450, a cheap p35 rig (either ip35e or ds3L if they're still around, maybe go p45 if p35 is gone by then) and 2x1gb ddr2. OC the hell out of it but keep voltages reasonable (nothing over 1.4vcore) and you'll get 3.4-3.6ghz. This will clobber the everliving crap out of most nehalems for a year or two, anyway, allowing the early-adopters to sort through all the bugs in nehalem v1.0.

Sound advice all the way around.

Compared to the OP's current rigs a 3.5'ish GHz Yorkfield will definitely be a speed demon for a couple years to come. Best to spend the extra dosh on video card, or invest it in something (bank CD, stocks, etc) that you know you can sell for more money later and then buy an upgrade with the proceeds.

 

[dmens]

Originally posted by: bryanW1995
and you'll get 3.4-3.6ghz. This will clobber the everliving crap out of most nehalems for a year or two

lol we'll see about that.

 

[Drsignguy]

Originally posted by: dmens

Originally posted by: bryanW1995
and you'll get 3.4-3.6ghz. This will clobber the everliving crap out of most nehalems for a year or two

lol we'll see about that.

This quote is very well put and way too funny:laugh:

 

[Nemesis 1]

Originally posted by: bryanW1995
It's fun watching the n00bs rip on the old hand...

Normally I'd say build a ddr3 c2d rig in august, but ddr2 is so unbelievably cheap that I wouldn't bother with it. ddr3 will drop WAY down when nehalem comes out b/c they won't even offer that in ddr2 iirc PLUS amd should have am3 out around the same time.

Having said that, I'd snag a Q9450, a cheap p35 rig (either ip35e or ds3L if they're still around, maybe go p45 if p35 is gone by then) and 2x1gb ddr2. OC the hell out of it but keep voltages reasonable (nothing over 1.4vcore) and you'll get 3.4-3.6ghz. This will clobber the everliving crap out of most nehalems for a year or two, anyway, allowing the early-adopters to sort through all the bugs in nehalem v1.0.

Do you have a link to that nehalem bench session . I really want to read it.

 

[demiurge3141]

Build cheap, and oc hard?I don't think there has ever been a better time for budget overclocking. Get some dirt cheap ddr2, an ip35-e and an e2140/60 and you will get like 100% oc for $150. This will do everything you want it to do, unless you fold, in which case spend $100 more on a q6600.

 

[Idontcare]

Originally posted by: Nemesis 1

Originally posted by: bryanW1995
It's fun watching the n00bs rip on the old hand...

Normally I'd say build a ddr3 c2d rig in august, but ddr2 is so unbelievably cheap that I wouldn't bother with it. ddr3 will drop WAY down when nehalem comes out b/c they won't even offer that in ddr2 iirc PLUS amd should have am3 out around the same time.

Having said that, I'd snag a Q9450, a cheap p35 rig (either ip35e or ds3L if they're still around, maybe go p45 if p35 is gone by then) and 2x1gb ddr2. OC the hell out of it but keep voltages reasonable (nothing over 1.4vcore) and you'll get 3.4-3.6ghz. This will clobber the everliving crap out of most nehalems for a year or two, anyway, allowing the early-adopters to sort through all the bugs in nehalem v1.0.

Do you have a link to that nehalem bench session . I really want to read it.

Everytime I think about Nehalem benches I just get day-dreamy eyes thinking about those 8 threads of goodness. Lord I hope it's affordable to build a desktop Nehalem system in 2009, I can't justify another $1500 CPU in today's economic climate.

 

[bryanW1995]

ok, let's step away from the brink for a minute. how much of a clock/clock advantage do you think that an affordable nehalem will have vs a comparbly priced penryn? Sure the high end will be nice, but I just finished reading a quote from an intel employee stating that they will not offer oc'ing on low/mid systems. Is it possible that they will? sure, it's possible, but even a low end 2.2 ghz nehalem will destroy anything that amd currently has or probably even will have in 09, so it does make sense for intel. Do any of you think that a 2.2 ghz nehalem will beat a 3.5 ghz penryn??? I'm personally very excited about nehalem, but with the lack of competition how long will intel wait to release it? I think that the high end will probably come out in q4 08 or q1 09, but I wouldn't be surprised at all to not see low/mid until summer 09 or later.

So, we have a choice of waiting/hoping vs a known top performer right now. How many of us have waited/hoped over the years for a better product to come out, only to be let down in the end? I'm glad NOW that I waited in july 07 to get a quad, but it took me 9 months instead of the 6 that I had planned for.


btw, based upon what the OP said, building a kickass 3.4-3.6 ghz 45nm quad system now would be perfect for his needs.

 

[Nemesis 1]

Originally posted by: bryanW1995
ok, let's step away from the brink for a minute. how much of a clock/clock advantage do you think that an affordable nehalem will have vs a comparbly priced penryn? Sure the high end will be nice, but I just finished reading a quote from an intel employee stating that they will not offer oc'ing on low/mid systems. Is it possible that they will? sure, it's possible, but even a low end 2.2 ghz nehalem will destroy anything that amd currently has or probably even will have in 09, so it does make sense for intel. Do any of you think that a 2.2 ghz nehalem will beat a 3.5 ghz penryn??? I'm personally very excited about nehalem, but with the lack of competition how long will intel wait to release it? I think that the high end will probably come out in q4 08 or q1 09, but I wouldn't be surprised at all to not see low/mid until summer 09 or later.

So, we have a choice of waiting/hoping vs a known top performer right now. How many of us have waited/hoped over the years for a better product to come out, only to be let down in the end? I'm glad NOW that I waited in july 07 to get a quad, but it took me 9 months instead of the 6 that I had planned for.


btw, based upon what the OP said, building a kickass 3.4-3.6 ghz 45nm quad system now would be perfect for his needs.

In the past I couldn't agree more. But now is a totally differant. The way I see it. This applies to all of no matter how we view performance PC s . Nehalem / Bulldozer should be long term investments . 5+ years. Heres what ya get from both systems.

1.) Fast single cores. 2 threads per core. 4-8 cores per die modular design.

2) Intel larrabbe more than likely will get hightest performance on an all intel system with Nehalem. Bulldozer same as intel best performance with ATI GPU.

3) DDR3 3 channel enough bandwidth for some time.

4) Intel SSD. expensive but pure performance . I would want 4x 80g raid 0 on pci-e.

5) 30" lcd

6) USB 3

This should keep both sides happy for a few years . LOL The Up grade I see is GPU and larrabbee 2 and ATI 800 what ever should make all happy for a couple of years.

 

[Jax Omen]

I honestly have no idea what the post above me is supposed to mean. Is that how he expects computing to be in 2009?

 

[bryanW1995]

you must speak nemesisese to understand him. sadly, I think I get what he means. allow me to feebly attempt a translation:

buy a high end intel system when nehalem comes out. it will be good for 5 years so don't worry about the $2,500+ price tag.

 

[aigomorla]

Originally posted by: rip
There's NO planning ahead. Decide in August.

+1

he said it as nice as it can be said.

We dont know what neha will do. We dont know what boards will come out for it. We dont know how the board will perform. We dont know how it will handle DDR3, we dont even know what the server side platforms look like yet.

So no we cant help you plan anything until intel decides to leak more info about it as well as the motherboard companies.

Anyone who is basing info off reviews, your sadly wrong. These are ES chips, and we all know you can only take ES chips with a grain of salt.

So dont confuse the OP in making choices. Who knows the low ball sector of neha might not be worth it over an overclocked Yorkfield.

So until we get more info, and i mean real info from a trusted review site, lets not make any assumptions.

 

[roguerower]

Originally posted by: bryanW1995
ok, let's step away from the brink for a minute. how much of a clock/clock advantage do you think that an affordable nehalem will have vs a comparbly priced penryn? Sure the high end will be nice, but I just finished reading a quote from an intel employee stating that they will not offer oc'ing on low/mid systems. Is it possible that they will? sure, it's possible, but even a low end 2.2 ghz nehalem will destroy anything that amd currently has or probably even will have in 09, so it does make sense for intel. Do any of you think that a 2.2 ghz nehalem will beat a 3.5 ghz penryn??? I'm personally very excited about nehalem, but with the lack of competition how long will intel wait to release it? I think that the high end will probably come out in q4 08 or q1 09, but I wouldn't be surprised at all to not see low/mid until summer 09 or later.

So, we have a choice of waiting/hoping vs a known top performer right now. How many of us have waited/hoped over the years for a better product to come out, only to be let down in the end? I'm glad NOW that I waited in july 07 to get a quad, but it took me 9 months instead of the 6 that I had planned for.


btw, based upon what the OP said, building a kickass 3.4-3.6 ghz 45nm quad system now would be perfect for his needs.

THANK YOU!!! This is exactly what I was looking for.

 

[semisonic9]

For what it's worth, I agree that you should build now.

Processors, mobos, and ram are all in a great place right now. You can OC your eyeballs out, if that's your thing, more safely and for less $$ than I've seen in recent times. Multi-GPU choices and mobo selection is the only part of the current market that drives me bananas, but I think ATI is going to make my choice much simpler in the near future with their 4xxx chips, supposedly out this month.

I'm not waiting. I'm pulling the trigger now.

~Semi

 

[Viditor]

Originally posted by: roguerower
I'm looking to build a new PC at the end of the summer, around the beginning of August. Now my problem is that I'm unsure about whether or not to go ahead with it or wait for Nehalem.

I understand that just b/c Nehalem comes out Core isn't going to disappear overnight, but I'd like to stay on top of the technology so I don't get screwed over.

My current plan was to build the system and then wait a year or so until the prices settled, then rip out the Mobo, CPU and RAM (possibly...heard it was going to DDR3) and then sell them and use the money to go with a new set.

Can anyone share their $0.02 on this and maybe tell me whether it's a stupid idea?

Nehalem desktops aren't due out till next year...
There will probably be an Extreme version late in Q4, but it won't be compatable to desktop sockets.

 

[Kuzi]

Originally posted by: dmens

Originally posted by: bryanW1995
and you'll get 3.4-3.6ghz. This will clobber the everliving crap out of most nehalems for a year or two

lol we'll see about that.

bryan1995 has a point especially when you keep cost into consideration. If Intel releases Nehalem in Q4 this year for Desktops, it'll probably be a high end Extreme Edition, costing $1000-$1500.

The Q9450 costs much less and will easily OC to 3.6GHz or more. Nehalem at release might beat that in some applications, but at what cost. Also Nehalem will most likely not OC as well.

 

[dmens]

Originally posted by: Kuzi
bryan1995 has a point especially when you keep cost into consideration. If Intel releases Nehalem in Q4 this year for Desktops, it'll probably be a high end Extreme Edition, costing $1000-$1500.

The Q9450 costs much less and will easily OC to 3.6GHz or more. Nehalem at release might beat that in some applications, but at what cost. Also Nehalem will most likely not OC as well.

um, buy a slower nehalem at the same price point and oc it? whats the point of comparing anything to a extremely expensive edition?

so the word is that nehalem will not overclock as well as penryn. where's the evidence and/or justification? i have not seen anything worth more than a pure WAG, all the rumors should be treated as such.

 

[Viditor]

Originally posted by: dmens

Originally posted by: Kuzi
bryan1995 has a point especially when you keep cost into consideration. If Intel releases Nehalem in Q4 this year for Desktops, it'll probably be a high end Extreme Edition, costing $1000-$1500.

The Q9450 costs much less and will easily OC to 3.6GHz or more. Nehalem at release might beat that in some applications, but at what cost. Also Nehalem will most likely not OC as well.

um, buy a slower nehalem at the same price point and oc it? whats the point of comparing anything to a extremely expensive edition?

so the word is that nehalem will not overclock as well as penryn. where's the evidence and/or justification? i have not seen anything worth more than a pure WAG, all the rumors should be treated as such.

Because that's the only desktop version being released this year...

 

[Kuzi]

I'm not sure, but I think lower clocked versions of Nehalem will not be released till next year. The OP asked if it's better to build a new system at the end of the summer, or wait for Nehalem. Don't know if he can wait till Q1 2009 for lower clocked Nehalem to be released.

There is no evidence that Nehalem will OC less than Penryn, we have to wait and see I guess, Q4 seems a long way off..

 

[Idontcare]

Originally posted by: dmens

Originally posted by: Kuzi
bryan1995 has a point especially when you keep cost into consideration. If Intel releases Nehalem in Q4 this year for Desktops, it'll probably be a high end Extreme Edition, costing $1000-$1500.

The Q9450 costs much less and will easily OC to 3.6GHz or more. Nehalem at release might beat that in some applications, but at what cost. Also Nehalem will most likely not OC as well.

um, buy a slower nehalem at the same price point and oc it? whats the point of comparing anything to a extremely expensive edition?

so the word is that nehalem will not overclock as well as penryn. where's the evidence and/or justification? i have not seen anything worth more than a pure WAG, all the rumors should be treated as such.

See Blauhung's post in the XS forums thread on 3.2GHz Nehalem: (Blauhung is an admitted Intel employee, but is not posting as an Intel rep obviously)

Originally posted by: Blauhung
Bloomfield and Gainstwon on the other hand have a scalable NB that will most likely be setting the base clock frequency that CPU and memory frequency. Just like now there will be selectable multipliers as well. Really the high end desktop and server platforms should be nice and configurable on the memory front.

So in summary, you should be fine with what you have, just don't expect to be able to tweak anything if you end up with the mainstream platform.

http://www.xtremesystems.org/f...=2961711&postcount=140

Sure its not the gospel, but Blauhung tends to be more right than wrong if you follow his statements and then see which ones come true months later.

But I am strongly inclined to believe Blauhung's version of Intel's planned future given that Intel MUST see an ROI for their efforts into courting the enthusiast crowd with SK and Bloomfield...leaving a Q6600'esque backdoor in their Nehalem profit plans would be a poor showing on the product managers behalf.