Corporate brain washing? I still say go for whichever is the best on the day you get the wallet out personally. The main issue is what upgrade routes will appear where in the i5/7/9 setup isn't clear in the long term and only time will tell on that front (even road maps change, they are intention not fact).
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Will Intel® Core™i7 (Bloomfield/X58) owners be able to afford an upgrade?
- Thread starter Big.Wayne
- Start date
Corporate brain washing? I still say go for whichever is the best on the day you get the wallet out personally. The main issue is what upgrade routes will appear where in the i5/7/9 setup isn't clear in the long term and only time will tell on that front (even road maps change, they are intention not fact).
This is pretty much the reason I went to Core i7 as well although rather ironically despite the high clocks achievable on my i7 920 I have never taken it past 3.2 ghz (yet) - to be fair I havn't decided on a aftermarket cooler yet.
The 6 core Gulftown processors do sound interesting but they will have to offer something exception for me (over my i7) to warrant upgrading.
Associate
- Joined
- 27 Feb 2007
- Posts
- 1,921
- Location
- Leeds
It depends what you mean by affordable. It's high end workstation/enthusiast, so it's never going to be cheap.
I doubt it's anything that sinister. Enthusiasts who are prepared to pay top dollar for X58, will more than likely be happy to pay top dollar again in the future. It's actually a benefit that Intel have confirmed X58 will be used beyond i7.
If Intel fleece enthusiasts too much with the lack of options and cost of i9, then enthusiasts will just end up just switching platforms. I doubt many would feel obliged to stay on x58 for the price of a mobo.
Well Xeon's have been on different sockets to Pentiums and Core 2's over the years. So rather than just server and mainstream, they've introduced a third option into the mix. Time will tell if they maintain this separation.
I just wondered what your thoughts were, I am looking at a new system and although the LGA1366/i7/Bloomfield is a stonking platform I am unclear what the processor upgrade path may look like?
Thanks in advance for any useful feedback!
You need to ask yourself what you're going to use the PC for, how long you want to keep it.
Even if you go x58 now, there's going to be upgrades to PCI/SATA etc, so by the time i9 arrives you might want a new x58 or at least 1366 (there may be a new chipset round the corner) based board anyway.
One thing that strikes me is that it is often overlooked that the way forward is increasing the number of cores instead of increasing clock speed. The high end platform will be the first to receive them, and using the Intel "tick tock" the mainstream platform will then receive them at the next refresh, for a time this will overlap the mainstream with the high end until the next high end release. Which is what we're seeing now with the i5 and i7 8xx's.
Ultimately everyone be it low end, mainstream, high end will get the benefits - but it's a question of how much you want to spend and how long you are prepared to wait for the that level of performance.
Last edited:
This could just be a subtle amalgamation of the server chips and the desktop chips. Maybe it will allow intel to eventually filter out all the crap sockets which must cost it money (lga 771 etc). Thus 1156 = mainstream, 1366 = very high end/server.
Associate
- Joined
- 1 Feb 2004
- Posts
- 426
- Location
- UK
Is it possible that these new ATI 5*** cards will saturate a single PCI-E x16 on the new 1156 motherboards? I just wish they would have had the new directX 11 cards out before they released the new 1156 socket. 
Off Topic:
I doubt it, if it did it would also saturate the PCI-E x16 slot on an X58 chipset also.
Ive never tried a xfire setup, but ive used sli on nforce 4 and 680i, ive decided to go x58 to give me the option of trying xfire or sli at x16 per slot, i had 7800gtx, 7900gts in sli and 8800gts 640's in sli before switching to a single 8800gts 512 then my current gtx 280, gonna pickup another 280 cheaply to give sli a blast on x58.
Associate
- Joined
- 17 May 2004
- Posts
- 1,585
Being honest with yourselves, by the time you NEED to replace an i7 920 overclocked, odds are you will want to replace MB for better sound and SATA-3 or 6Gbps anyway.
Some of the P55 boards are 3 x x16 physically but inly x8, x8 and x4 electrically. The P45 provided 2 x x16 physically but x8, x8 electrically so I don't see a change there.
Buying a system for an upgrade path is stupid, by the time you have filled in the payment details it feels like it is out of date already.
Some of the P55 boards are 3 x x16 physically but inly x8, x8 and x4 electrically. The P45 provided 2 x x16 physically but x8, x8 electrically so I don't see a change there.
Buying a system for an upgrade path is stupid, by the time you have filled in the payment details it feels like it is out of date already.
He thanks for that setter . . .of course no one mentioned SLI or Crossfire so you (& hardtarget) are in fact pretty much Off Topic! 
What are your thoughts about the future cost of LGA1366 chips? you don't seem shy to spend a few hundred quid . . . if 6 months down the line your a bit bored with your 920 would you be prepared to shell out several hundred pounds on a Gulftown? (if it goes that way).
What are your thoughts about the future cost of LGA1366 chips? you don't seem shy to spend a few hundred quid . . . if 6 months down the line your a bit bored with your 920 would you be prepared to shell out several hundred pounds on a Gulftown? (if it goes that way).
Last edited:
Yep i probably would mate, ive spent far more money on things that i never use much theese days, shooting clay pigeons is another hobby of mine, i bought a beretta shotgun at £2000.00, got it fitted to me by a professional coach, i dont even shoot much now. my idea with lga 1366 is to try it out, ive read your post on the upgrade path of this socket (good read) and some very good points about the future of this socket, my nephew has got a nice we lynnfield rig up and running this week which i must go and see in action, as a gamer (1920x1200) the lynnfield is a bit crippled at x8 per lane with dual cards. my current a8n32 sli board will do x16 per lane and its an old s939 board.
I think maybe you are seeing things differently to me. It should be obvious there is no NEED for a lot of upgrades we do however we do them anyway for out own good reasons. Just think of Women with their bags and handbags, of course they don't actually NEED several dozen of each but thats what always seems to happen!
I think a large number of forum members just like to play with the hardware, swap bits about etc, I think of it as mechano for grown up kids, the big difference being that these days a lot of the pieces don't fit together, so it could be kinda a drag if your not able to Pop & Swap chips like the old days!
I think you could have worded that better!
The opposite end of the spectrum to your last comment would be buying a sealed box, its got hardware in it, you don't really know what it is but its a computer and it does stuff.
Us hardware enthusiasts generally always consider the route ahead when buying stuff, not all but most I would say!
Some very good points wayne, i tend to upgrade graphics cards pretty regularly, gaming at 1920x1200 doesnt help, also with cpu's lately, i had a q6600 at 3.8ghz (to power hungry) then a q9550 (sold) now waiting on i7, i do tend to get sucked into the upgrade craze, a while back i seen a blu ray drive on the mm and the old "i want that kicked in", great purchase though.
I think people are viewing £500 six core chips in the wrong light. A gamer does not want 12 threads, they're only just managing to justify four. There are precious few things that can use so many processing cores. I am choosing to ignore encoding, because I just can't believe it matters whether something that can be run overnight takes 6 hours or 8.
However, where there is a market for this is workstations. CAD (some anyway) loves ram, processors and bandwidth. Enter the six core X58 chip. Servers have always loved ram and processors. Virtual machines will also love this, so there are options like running a really large number of thin clients off one X58 board.
Previously these roles were filled by dual (or quad) processor boards. So, now you would be using two xeon chips and a load of ecc ram. However, these cost a lot of money. It's of the order of £1500 for board, two nehalem chips, and 12gb of ram. Alternatively you can have one quad core, ram and board for £500. So this six core chip would come in at 600+ram+board at around £900. So it represents a 'budget' workstation build. I think it is priced absolutely spot on, coming between single and dual socket systems, assuming it is around £600 and works in current x58 boards (big assumptions).
Finally, a 4.2ghz i7 seems to be doing alright against two nehalem chips at stock. So an overclocked six core one, say at 3.8ghz, is likely to storm all over a dual socket board which can't overclock.
As for whether I'll be able to get one, I don't know. The processor is the most important part of the machine in my eyes. So if I can find the funds I will probably get one. What will convince me otherwise is if the software I'm using at the time wont use more cores, I'm currently running solid edge which only uses two anyway. Or if the price turns out to be well out of reach / they don't clock worth a damn.
Cheers for the thread
However, where there is a market for this is workstations. CAD (some anyway) loves ram, processors and bandwidth. Enter the six core X58 chip. Servers have always loved ram and processors. Virtual machines will also love this, so there are options like running a really large number of thin clients off one X58 board.
Previously these roles were filled by dual (or quad) processor boards. So, now you would be using two xeon chips and a load of ecc ram. However, these cost a lot of money. It's of the order of £1500 for board, two nehalem chips, and 12gb of ram. Alternatively you can have one quad core, ram and board for £500. So this six core chip would come in at 600+ram+board at around £900. So it represents a 'budget' workstation build. I think it is priced absolutely spot on, coming between single and dual socket systems, assuming it is around £600 and works in current x58 boards (big assumptions).
Finally, a 4.2ghz i7 seems to be doing alright against two nehalem chips at stock. So an overclocked six core one, say at 3.8ghz, is likely to storm all over a dual socket board which can't overclock.
As for whether I'll be able to get one, I don't know. The processor is the most important part of the machine in my eyes. So if I can find the funds I will probably get one. What will convince me otherwise is if the software I'm using at the time wont use more cores, I'm currently running solid edge which only uses two anyway. Or if the price turns out to be well out of reach / they don't clock worth a damn.
Cheers for the thread
Associate
- Joined
- 1 Feb 2004
- Posts
- 426
- Location
- UK
Sorry Wayne I did stray. 
Excellent thread and some very good points being made. I myself have to choose which path to go X58 or P55 either way I like to see clear upgrade path and from reading your thread I just don't see one..........well not if I choose X58 we still don't know for sure if the future LGA1366 chips will be compatible with the current socket and then theres the cost's involved so many unanswered questions and I need to make a decision soon not sure how much longer I can go on using this old system sold most of my signature system thinking it was a simple move upto 1366 
Excellent thread and some very good points being made. I myself have to choose which path to go X58 or P55 either way I like to see clear upgrade path and from reading your thread I just don't see one..........well not if I choose X58 we still don't know for sure if the future LGA1366 chips will be compatible with the current socket and then theres the cost's involved so many unanswered questions and I need to make a decision soon not sure how much longer I can go on using this old system sold most of my signature system thinking it was a simple move upto 1366 Best advice i could give you ,sit and think what you want your pc for,ignore what evryone else has,because a lot of ppl brought the i7920 due to its hype and bragging rights,but most wil never use the full potentinal of it,as for the future upgrade path thats really hard to figure out,we know the i9 will be coming but unless you got £1000 just lying around its going to be out of reach for most ppl,as for the 1156 its going to be around for a long time and there will be better chips etc.and as the others have said do you really need more cores for gaming,i don,t think so quad cores have been out for some yrs now and still they have never been pushed to the limit..go with what fits your needs instead of following the crowd. 
gulftown...
will be a good upgrade for workstations, not really for the average user
[tbh i7 is more than enough for the average user]
in an environment where time is money - getting that bit more done will be just the ticket
im not going to discuss too much but it is going to be very impressive
i think however that it would be the equivalent of when we all went from 65 to 45 nm quads - the gaines were obvious to be seen, but did they really offer much of a benefit over the older counterparts?
chances are i will get one to see what it can do [but thats only because i cant help fiddling with hardware]
personally i am seeing far greater returns in upgrading my storage solutions, i think for most x58/p55/790xt owners - that is where there money is best spent currently
great thread btw wayne...
will be a good upgrade for workstations, not really for the average user
[tbh i7 is more than enough for the average user]
in an environment where time is money - getting that bit more done will be just the ticket
im not going to discuss too much but it is going to be very impressive
i think however that it would be the equivalent of when we all went from 65 to 45 nm quads - the gaines were obvious to be seen, but did they really offer much of a benefit over the older counterparts?
chances are i will get one to see what it can do [but thats only because i cant help fiddling with hardware]
personally i am seeing far greater returns in upgrading my storage solutions, i think for most x58/p55/790xt owners - that is where there money is best spent currently
great thread btw wayne...
Associate
- Joined
- 17 Sep 2009
- Posts
- 546
- Location
- UK
for anyone doing 3d rendering like myself this chip will be used to its full potential. going from a heavily OC'd quad to an i7 gave me about a 1.5x boost in rendering speed at least (in all 3d applications).
I probably will not buy an i9 but if anyone is in the CAD/3D pipeline and has the money i'm sure it will be beneficial...but not necessarily money well spent because its still a little overpriced for what it is.
I probably will not buy an i9 but if anyone is in the CAD/3D pipeline and has the money i'm sure it will be beneficial...but not necessarily money well spent because its still a little overpriced for what it is.
Last edited: