Originally Posted by article ...we promised Intel that we wouldn't reveal any performance results just yet...
Did Intel say when this could happen?
That's a really neat looking board, I love the lack of a huge heatpipe assembly!
The new mounting holes for cooling is ridiculously dumb though.. Intel stands to gain nothing here, two more MM and nobody would have any problems. Pretty much every cooling company already has i7 heatsinks out, the change in form factor is just dumb.
You couldn't give us one graph or table? How does it look on the test bench?
Meh. Hardly a 5 Ghz overclock on air like they were waving some time ago, I seriously doubt you had 3.8 Ghz stock speed cpu.
Real i7 still looks like more desirable option, unless the platform costs are significantly, like atleast 30% lower. Castrating the SLI/Crossfire by limiting the amount of PCIe-lanes was plain retarded.
Less PCI-E lanes isn't much of bottleneck yet, maybe DX11 cards will change that, but there's no way to know.
I'll stay single-card, just like almost everybody. P55 is more mainstream, dual x16 PCI-E isn't important to the mainstream crowd. If you need that, stick with X58.
Originally Posted by SlowMotionSuicide Meh. Hardly a 5 Ghz overclock on air like they were waving some time ago, I seriously doubt you had 3.8 Ghz stock speed cpu.
To be fair, the article does say 'at least'... dependent on whether you're an optimist or a pessimist, you can decide whether that means, "We got 1.21GHz overclock before it wouldn't even POST any more" or "We got it to a 100% overclock with no sweat but we promised Intel (read: we're under NDA) that we wouldn't brag about it". ;)
If that layout remains the same, the P55 chip overheating is going to be a fairly common occurrence. No heatpipes is a good thing for watercoolers/aftermarket cooling improvements... but I can't believe that Asus haven't spotted a potential issue with cooling the P55.
Are these articles just being taken verbatim from CPC? I've seen this in a few articles now and its rather annoying. Its usually a reference to a story which has already been on bit-tech so its not as if you couldn't link them. And who is editing these articles such that they get through with references to a different publication?
Originally Posted by Cobalt Are these articles just being taken verbatim from CPC? I've seen this in a few articles now and its rather annoying. Its usually a reference to a story which has already been on bit-tech so its not as if you couldn't link them. And who is editing these articles such that they get through with references to a different publication?
I'm not going to name and shame - instead I'll blame it on being a Monday morning :p
Originally Posted by Cobalt Are these articles just being taken verbatim from CPC....who is editing these articles such that they get through with references to a different publication?
CPC and Bit-Tech, AFAIK, same family and same writers in some instances.
As CPC website no longer seems to appear to have anything, and I do wonder why it even appears on the web seeing as it is never updated, it's no surprise some of the CPC info now shows up on Bit Tech.
Originally Posted by SlowMotionSuicide Meh. Hardly a 5 Ghz overclock on air like they were waving some time ago, I seriously doubt you had 3.8 Ghz stock speed cpu.
To be fair, the article does say 'at least'... dependent on whether you're an optimist or a pessimist, you can decide whether that means, "We got 1.21GHz overclock before it wouldn't even POST any more" or "We got it to a 100% overclock with no sweat but we promised Intel (read: we're under NDA) that we wouldn't brag about it". ;)
If that layout remains the same, the P55 chip overheating is going to be a fairly common occurrence. No heatpipes is a good thing for watercoolers/aftermarket cooling improvements... but I can't believe that Asus haven't spotted a potential issue with cooling the P55.
Yep, Intel wouldn't want news that these chips overclock well spreading into the press. They have to make sure that everyone is quiet about how good these chips are, or else people might buy them.
Originally Posted by TurtlePerson2 Yep, Intel wouldn't want news that these chips overclock well spreading into the press. They have to make sure that everyone is quiet about how good these chips are, or else people might buy them.
I think they don't want people to see how good they are and decide to wait for them instead of buying the more expensive i7s now.
Originally Posted by perplekksParadigm Shifter If that layout remains the same, the P55 chip overheating is going to be a fairly common occurrence.
I'd imagine the overheating P55 chip will be sorted out by time the boards are in mass production - the board we had sported a beta P55 chip (i.e. non-final silicon)
Quote:
Originally Posted by perplekks45 Looks interesting and the last paragraph is mouth-watering. Any hints on when we can possibly expect a full review? Maybe with prices even?
That all depends on Intel, but I'd be surprised if we'll be allowed to show any of the awesome performance results we've already got for a few more weeks.
Originally Posted by docodine The new mounting holes for cooling is ridiculously dumb though.. Intel stands to gain nothing here, two more MM and nobody would have any problems.
I initially thought the same but then I remembered when they updated the Xeon's from 7520 to 5000 chipset. Same mounting holes, but the proc was about a millimeter taller or less tall or something. So if you used an old heatsink it seemed to install & looked ok, but would actually overheat pretty quickly because it didn't make proper contact.
So if with Lynfield for some reason they couldn't keep the proc at the same thickness, it actually makes sense to have different mounting holes, as annoying as it is!
Originally Posted by Lizard I'd imagine the overheating P55 chip will be sorted out by time the boards are in mass production - the board we had sported a beta P55 chip (i.e. non-final silicon
I would certainly hope so. However, forgive me for being sceptical, but while I'd accept that the board layout/heatsink size might change... if the chip is overheating now, I doubt that it'll suddenly stop overheating just because Intel finalise the silicon.
BAH! DAMN YOU BIT-TECH! You make me so sad T_T
All I want to know about right now is the price and performance of Lynnfield in comparison to AM3 and what do you do? You do a P55 Mobo preview which looks sexy as hell and makes me want to know even more T_T
All I have to say is now you've made me check my RSS feed every 15 minutes for news/reviews of the Lynnfield processors which I know won't be there for weeks, the worlds a cruel place indeed :(
EDIT: One thing I'd like to know, does it use as much power as i7? Overclock and no Overclock? I'm not sure they'd care about you telling us that info but if you can't I'll live.
I seriously doubt that 5 GHz is possible on air, given the fact that even the most expensive Lynnfield is only clocked at just under 3 GHz, and that is a $500 CPU. Still, a 4 GHz overclock with hyper-threading is not to be poo-pooed, it just does raise the benchmark any higher than the i7.
I hope that a Lynnfield's total build price is significantly lower than an equivalent i7, it seems to me that given the monthly reduction in price of i7 parts, a future Lynnfield build will not actually represent any better value, than a time-adjusted i7 one.
Will Lynnfield get a die shrink to 32nm in the future, or just give way to Sandy Bridge?
...using the latter socket support cheaper dual-channel DDR3 rather than triple-channel DDR3
Feel free to correct me if I am wrong, but I was under the impression that how many channels RAM ran with was dependent entirely on the chipset, and not on the RAM? If you happen to have 3 2GB sticks of RAM, having used 2 of them in dual channel mode previously, sticking them all into a triple channel capable board would provide just the same performance as buying a new kit of 3 sticks (assuming all else is equal of course).
Originally Posted by Skiddywinks Feel free to correct me if I am wrong, but I was under the impression that how many channels RAM ran with was dependent entirely on the chipset, and not on the RAM? If you happen to have 3 2GB sticks of RAM, having used 2 of them in dual channel mode previously, sticking them all into a triple channel capable board would provide just the same performance as buying a new kit of 3 sticks (assuming all else is equal of course).
My understanding is that on the lga 1366, you have to use higher density ram to run large amounts of memory e.g. 12 GB in triple channel, would mean 3 x 4 GB ram sticks. In theory, on 1156 boards, you can use two pairs of lower density ram to achieve similar results, lower density ram sticks are of course cheaper.
Originally Posted by Skiddywinks Feel free to correct me if I am wrong, but I was under the impression that how many channels RAM ran with was dependent entirely on the chipset, and not on the RAM? If you happen to have 3 2GB sticks of RAM, having used 2 of them in dual channel mode previously, sticking them all into a triple channel capable board would provide just the same performance as buying a new kit of 3 sticks (assuming all else is equal of course).
My understanding is that on the lga 1366, you have to use higher density ram to run large amounts of memory e.g. 12 GB in triple channel, would mean 3 x 4 GB ram sticks. In theory, on 1156 boards, you can use two pairs of lower density ram to achieve similar results, lower density ram sticks are of course cheaper.
I'm not sure I follow. All LGA1366 boards I know of have 6 DDR slots, allowing up to 12GB or RAM fairly easily. Of course, since it is a triple channel chipset, you want multiples of three for your RAM (hence, 6 slots). LGA775 and LGA1156 boards are dual channel, so thus have 4 slots (and thus 8GB fairly easily).
RAM density is a moot point here as you can use whatever density you want on either chipset/board. The only difference between the boards is the number of channels (and as a knock on effect, number of slots). Given that, it is infact harder/more expensive to get 12GB on LGA775 and 1156 since there are less slots.
My point still stands, and after a check on wikipedia it seems I am correct; the number of channels RAM runs in/on is dependent on the board entirely.
Originally Posted by Skiddywinks My point still stands, and after a check on wikipedia it seems I am correct; the number of channels RAM runs in/on is dependent on the board entirely.
It's dependent on the memory controller. Which in the case of i7 is on the CPU die.
Looks to be a nice board if the price is right. The slot layout is exactly what I need too. :) But I'm lost for words why the IDE connector is given prime real-estate ahead of the SATA ports? And a COM port? Madness.
Be interesting what the Deluxe brings to the table. Hard to pin down considering what we know about the M3F.
Comments 1 to 25 of 34
ReplyDid Intel say when this could happen?
That's a really neat looking board, I love the lack of a huge heatpipe assembly!
The new mounting holes for cooling is ridiculously dumb though.. Intel stands to gain nothing here, two more MM and nobody would have any problems. Pretty much every cooling company already has i7 heatsinks out, the change in form factor is just dumb.
You couldn't give us one graph or table? How does it look on the test bench?
Real i7 still looks like more desirable option, unless the platform costs are significantly, like atleast 30% lower. Castrating the SLI/Crossfire by limiting the amount of PCIe-lanes was plain retarded.
I'll stay single-card, just like almost everybody. P55 is more mainstream, dual x16 PCI-E isn't important to the mainstream crowd. If you need that, stick with X58.
page 15 ??? what page 15 ??
If that layout remains the same, the P55 chip overheating is going to be a fairly common occurrence. No heatpipes is a good thing for watercoolers/aftermarket cooling improvements... but I can't believe that Asus haven't spotted a potential issue with cooling the P55.
I'm not going to name and shame - instead I'll blame it on being a Monday morning :p
Fixed now.
As CPC website no longer seems to appear to have anything, and I do wonder why it even appears on the web seeing as it is never updated, it's no surprise some of the CPC info now shows up on Bit Tech.
Yep, Intel wouldn't want news that these chips overclock well spreading into the press. They have to make sure that everyone is quiet about how good these chips are, or else people might buy them.
I'd imagine the overheating P55 chip will be sorted out by time the boards are in mass production - the board we had sported a beta P55 chip (i.e. non-final silicon)
That all depends on Intel, but I'd be surprised if we'll be allowed to show any of the awesome performance results we've already got for a few more weeks.
I initially thought the same but then I remembered when they updated the Xeon's from 7520 to 5000 chipset. Same mounting holes, but the proc was about a millimeter taller or less tall or something. So if you used an old heatsink it seemed to install & looked ok, but would actually overheat pretty quickly because it didn't make proper contact.
So if with Lynfield for some reason they couldn't keep the proc at the same thickness, it actually makes sense to have different mounting holes, as annoying as it is!
All I want to know about right now is the price and performance of Lynnfield in comparison to AM3 and what do you do? You do a P55 Mobo preview which looks sexy as hell and makes me want to know even more T_T
All I have to say is now you've made me check my RSS feed every 15 minutes for news/reviews of the Lynnfield processors which I know won't be there for weeks, the worlds a cruel place indeed :(
EDIT: One thing I'd like to know, does it use as much power as i7? Overclock and no Overclock? I'm not sure they'd care about you telling us that info but if you can't I'll live.
I hope that a Lynnfield's total build price is significantly lower than an equivalent i7, it seems to me that given the monthly reduction in price of i7 parts, a future Lynnfield build will not actually represent any better value, than a time-adjusted i7 one.
Will Lynnfield get a die shrink to 32nm in the future, or just give way to Sandy Bridge?
Feel free to correct me if I am wrong, but I was under the impression that how many channels RAM ran with was dependent entirely on the chipset, and not on the RAM? If you happen to have 3 2GB sticks of RAM, having used 2 of them in dual channel mode previously, sticking them all into a triple channel capable board would provide just the same performance as buying a new kit of 3 sticks (assuming all else is equal of course).
My understanding is that on the lga 1366, you have to use higher density ram to run large amounts of memory e.g. 12 GB in triple channel, would mean 3 x 4 GB ram sticks. In theory, on 1156 boards, you can use two pairs of lower density ram to achieve similar results, lower density ram sticks are of course cheaper.
I'm not sure I follow. All LGA1366 boards I know of have 6 DDR slots, allowing up to 12GB or RAM fairly easily. Of course, since it is a triple channel chipset, you want multiples of three for your RAM (hence, 6 slots). LGA775 and LGA1156 boards are dual channel, so thus have 4 slots (and thus 8GB fairly easily).
RAM density is a moot point here as you can use whatever density you want on either chipset/board. The only difference between the boards is the number of channels (and as a knock on effect, number of slots). Given that, it is infact harder/more expensive to get 12GB on LGA775 and 1156 since there are less slots.
My point still stands, and after a check on wikipedia it seems I am correct; the number of channels RAM runs in/on is dependent on the board entirely.
It's dependent on the memory controller. Which in the case of i7 is on the CPU die.
Be interesting what the Deluxe brings to the table. Hard to pin down considering what we know about the M3F.
-
« Previous
-
1
-
2
-
Next »
Discuss in the forums