Core i7 wake up call: AMD Phenom II X6 1090T BE overclocked to 6.29GHz!!

[DrMrLordX]

No it doesn't. You are only comparing average framerates. When comparing high end processors, the focus should be on minimum framerates, not just averages, because that's where the architectural differences are most prevalent in contributing to a smooth gaming experience. If you just look at average framerates, then you are more GPU limited since more or less every modern CPU can provide sufficient averages.

So Clarkdale is going to deliver better minimum framerates . . . why exactly?

Besides, why in the heck are we trying to figure out how Bloomfield will fare against Thuban by comparing Clarkdale to Deneb? Bloomfield and Clarkdale have very little in common aside from basic core architecture, and Deneb has a 100% advantage in physical core count over Clarkdale while Thuban only has a 50% advantage in physical core count over Bloomfield.

Phenom x4 965be is clocked @ 3.4ghz while i3 is clocked @ 2.93 ghz. If you want to compare at least compare at same clock levels or even same price range. At least that's what I've been told by guys in this thread who favor AMD. It's hypocritical. :hmm:

Eh wot? The 3.4 ghz 965BE beats every Clarkdale in 6 out of 8 benches, regardless of clockspeed. That includes the i5-661 at 3.33 ghz.

Face it, Clarkdale just isn't the best gaming CPU out there, and with Thuban/Zosma launching, there's practically no reason to mess with one for games. In fact, there's no reason right now. Why are we talking about Clarkdale again?

 

[RussianSensation]

Yeah lets all focus on those intangible gains, that's what counts here.

You realize AVG framerate is function of the maximum and minimum framerates as well? So you can have a game that has less demanding areas where the framerate shoots to 120fps (which has little to no benefit over 60), but then if it dips to 15fps, it's unplayable. You can still have an outstanding 60fps average in this game but even having 3-4 times a minute when framerate tanks below 20fps results in a dissatisfactory gameplay experience. If you don't think minimum framerates are important, why didn't you just buy a console and play at 30fps?

Intangible? You are saying that if in 1 minute benchmarking, a game drops to 15 fps 3x for 0.1 seconds while still producing 60fps avg is not a big deal? That's exactly what reduces playability. One of the most important areas where a CPU benefits in gaming is minimum framerates. Everyone just chooses to ignore it because "OH a slight drop for a millisecond is not important because it's not constant...". This is exactly what produces random choppiness.

... your tests are a massive waste of time being gpu limited. Identical architecture with common sense results, were you just bored or something?

He provided interesting real world data. What did you provide? Like seriously how did you add any value in this thread at all? If you have nothing concrete to provide, at the very least you can be respectful to members on this forum.

 

[BenchZowner]

 

[3DVagabond]

We don't but there are plenty of benches out there that compare the i3 vs x4 and phenom x4. i3 does lose but not much even in these apps. That's 100% more core though. x6 is only 50% more core over i7. It would be very close.

As for games even in 4 threaded games i3 very competitive to phenom x4. It's not that hard to find benches. :hmm:

It's real simple. If you are playing games you don't need 6 cores. Some people though actually use their computers for computing. Those people will appreciate (and use) 6 core processors at $300.00 and less, rather than at $1000.00. I'm not saying that the $300.00 six core is as good as the $1000.00 six core. I can't afford the $1000.00 one though and I can afford the $300.00 one. See the logic?

 

[Nemesis 1]

So Clarkdale is going to deliver better minimum framerates . . . why exactly?

Besides, why in the heck are we trying to figure out how Bloomfield will fare against Thuban by comparing Clarkdale to Deneb? Bloomfield and Clarkdale have very little in common aside from basic core architecture, and Deneb has a 100% advantage in physical core count over Clarkdale while Thuban only has a 50% advantage in physical core count over Bloomfield.



Eh wot? The 3.4 ghz 965BE beats every Clarkdale in 6 out of 8 benches, regardless of clockspeed. That includes the i5-661 at 3.33 ghz.

Face it, Clarkdale just isn't the best gaming CPU out there, and with Thuban/Zosma launching, there's practically no reason to mess with one for games. In fact, there's no reason right now. Why are we talking about Clarkdale again?

Not if I handpick the Benchmarks it doesn't . Look AMD cpus are OK good performance .

But it falls short of intel cpus Clock for clock i3 beats anything AMD has in the majority of ALL benchmarks . Now if we hand pick ya your correct in some benchies . All of the sudden single threaded benchmarks don't count . Dualthreaded benchies don't count . You guys are reaching .

 

[Nemesis 1]

So Clarkdale is going to deliver better minimum framerates . . . why exactly?

Besides, why in the heck are we trying to figure out how Bloomfield will fare against Thuban by comparing Clarkdale to Deneb? Bloomfield and Clarkdale have very little in common aside from basic core architecture, and Deneb has a 100% advantage in physical core count over Clarkdale while Thuban only has a 50% advantage in physical core count over Bloomfield.



Eh wot? The 3.4 ghz 965BE beats every Clarkdale in 6 out of 8 benches, regardless of clockspeed. That includes the i5-661 at 3.33 ghz.

Face it, Clarkdale just isn't the best gaming CPU out there, and with Thuban/Zosma launching, there's practically no reason to mess with one for games. In fact, there's no reason right now. Why are we talking about Clarkdale again?

Lets do this but lets do it fair and make it payoff . NO cookies .

My i5 670 on water against any 4 core amd processor . Highest clocks each will attain on water. Lets say a brand new 6 core AMD as the bet.

 

[BD231]

He provided interesting real world data. What did you provide? Like seriously how did you add any value in this thread at all? If you have nothing concrete to provide, at the very least you can be respectful to members on this forum.

I don't contribute to threads that have nothing to offer.

 

[Apocalypse23]

Not if I handpick the Benchmarks it doesn't . Look AMD cpus are OK good performance .

But it falls short of intel cpus Clock for clock i3 beats anything AMD has in the majority of ALL benchmarks . Now if we hand pick ya your correct in some benchies . All of the sudden single threaded benchmarks don't count . Dualthreaded benchies don't count . You guys are reaching .

I think what we all need to compare is the price to performance ratio here, and not so much clock per clock, perhaps even weigh in both those aspects together and get the best of both worlds.

Let's say we want optimal performance with a future-proof system on both i7s and thubans, and let's pick some of the highest parts for each setup, well in this case:

The i7 930/920 is $300 (selling for $200-289 at some stores), but will require an X58 board ($200 for an avg board) and DDR3 (triple channel for optimal performance).
Pros:4 core + 8 threads + Turbo Boost

The Thuban 1055/1090T will be selling for $200-$300, but will require an AM3 (lets say we want to ditch older boards) board ($140 for a decent board) and lets say DDR3 (dual channel).
Pros:6 cores + turbo Core

If you look at these comparisons, the only advantage AMD has is not having to go triple channel and will just need dual channel for optimal efficiency. Also, you end up saving about $60 dollars avg if going with a AM3 board. So the board and RAM will save you money when upgrading with AMD's thubans.

I think the fact that Intel actually offers Hyperthreading (8 cores!) will make up for the lack of the additional 2 cores and as a result it will directly compete with AMD's Thuban, since we already know that the thuban's are slower with the 4 cores. I think it's a fair comparison with 6 cores. Also, we can't expect a competitor to offer a product that exactly resembles the same architecture or cores of it's opponent, it could simply represent a competitive chip that performs close or equal to the competition in different ways: be it 6 core or 12 core.

We also can't say that we need to compare two different chips on a clock for clock basis as both are extremely different architectures and technologies. What we can compare is the price and that is the bottom line. If the AMD chip offered Hyperthreading with the 6 core chips and still performed similar to the i7 930/920, then it would be somewhat disappointing, but we have a chip that offers no HT and makes up for the difference with 6 cores and a high OC ability, and then things level off with whatever it takes. So in this case, maybe a 1090T oced at 4.5Ghz on air would equal an i7 930 oced at 4-4.2ghz? But the 1090T makes up for this by offering higher OC as opposed to the i7 930 on air (speculating).

If my instincts are correct, then the 1090T BE will be the special Overclocker's chip which should be able to run on average at a mid point between the i7 930 and the i7 980X in terms of overall performance.

 

[DrMrLordX]

Not if I handpick the Benchmarks it doesn't . Look AMD cpus are OK good performance . .

Handpick? What are you going on about? Those were Anandtech's benchmarks. These are Anandtech's forums. AzN said i3 was better in games . . . and, at least according to the benchmarks carried out by the people that host these forums, he was wrong.

Anand can't win. Go to AMDzone and they all think he's been bought and paid for by Intel because of an allegedly-bogus AES benchmark in Johan's Magny-Cours article. Then you show up, on these forums, and claim that someone linking Anand's benchmarks is "handpicking" results.

As far as watercooled i5-670s go . . . how about we just say no and forget about that? Or do I have to point out that an i5-750 is cheaper and faster than the i5-670? Why would anyone care about a 670 flailing around at 4.7 ghz, trying to beat someone's watercooled 965BE when they could trash any Deneb (or Clarkdale) with a processor that costs $100 less than the 670?

If that doesn't highlight just how ridiculous Clarkdale is as a performance processor (except for SuperPi 1M runs), then I don't know what does.

Look, if you want to take that i5-670 and square off against some 4.5 ghz Deneb on water you go right ahead, I want no part of that. Someone with an air-cooled 750 will beat them both on a majority of benchmarks. And the 750 costs $100 less retail than the 670.

And again, why is anyone even bringing up Clarkdale in this thread? It makes no sense.

It's real simple. If you are playing games you don't need 6 cores.

And then there was light, and it was good.

 

[heyheybooboo]

When did Tom's hijack the Anand CPU forum?






--

 

[Cogman]

You realize AVG framerate is function of the maximum and minimum framerates as well? So you can have a game that has less demanding areas where the framerate shoots to 120fps (which has little to no benefit over 60), but then if it dips to 15fps, it's unplayable. You can still have an outstanding 60fps average in this game but even having 3-4 times a minute when framerate tanks below 20fps results in a dissatisfactory gameplay experience. If you don't think minimum framerates are important, why didn't you just buy a console and play at 30fps?

You realize that minimum frame rate isn't an average, it is a one time occurrence? You realize that average, while it takes into account the minimum frame rate, is hardly affected by it?

Minimum frame rates doesn't tell you how often the minimum frame rate happens, nor does it tell you anything particularly useful about the CPU. Its about as useful as someone reporting a CPU goes to 6.29 GHz on liquid nitrogen.

Intangible? You are saying that if in 1 minute benchmarking, a game drops to 15 fps 3x for 0.1 seconds while still producing 60fps avg is not a big deal? That's exactly what reduces playability. One of the most important areas where a CPU benefits in gaming is minimum framerates. Everyone just chooses to ignore it because "OH a slight drop for a millisecond is not important because it's not constant...". This is exactly what produces random choppiness.

And how exactly does the minimum frame rate report that information? Here's a tip, it doesn't. It reports the minimum for the system. Who's to say that the .1 ms 15 fps dip happens more then once in the game? What if that is a one time only thing and the rest of the dips are only going down to 30fps. Then what? You'll condemn a CPU just because it doesn't stand up to your arbitrary requirement?

He provided interesting real world data. What did you provide? Like seriously how did you add any value in this thread at all? If you have nothing concrete to provide, at the very least you can be respectful to members on this forum.

As I've said and posted links to, his tests ARE worthless. I posted links to a couple of well trusted and known review sites that show that people that really test components don't look at Minimum FPS (one of those sites being anandtech.) for exactly the reasons I've been saying (without refute from benchzowner)

However, I think I'll follow benchzowner's advice and

Though, I don't think anyone in this thread has been trolling.

 

[AzN]

Eh wot? The 3.4 ghz 965BE beats every Clarkdale in 6 out of 8 benches, regardless of clockspeed. That includes the i5-661 at 3.33 ghz.

Face it, Clarkdale just isn't the best gaming CPU out there, and with Thuban/Zosma launching, there's practically no reason to mess with one for games. In fact, there's no reason right now. Why are we talking about Clarkdale again?

Because clarkdale paints a picture how Phenom II x6 is going to compare to i7 860 for instance. Again. You rather compare a $110 processor with a $180 processor. 2 intel core vs 4 AMD core CPU's. I understand i3 is slightly slower than Phenom when comparing @ stock but clock for clock it's very close even in quad optimized games but when overclocked on both ends i3 beats it in more games than a phenom x4. If an i3 with 100% less core is competitive with AMD's best at the moment what makes you think AMD's 6 core CPU is going to compare with intel's i7 4 core with only 50% less core?

 

[AzN]

It's real simple. If you are playing games you don't need 6 cores. Some people though actually use their computers for computing. Those people will appreciate (and use) 6 core processors at $300.00 and less, rather than at $1000.00. I'm not saying that the $300.00 six core is as good as the $1000.00 six core. I can't afford the $1000.00 one though and I can afford the $300.00 one. See the logic?

You need 4 cores at the moment at least starting to pickup where dual core was at it's infancy of being mainstream optimized. Even in quad optimized games i3 easily keep paces with AMD's best quad.

I'm just trying to paint a picture here comparing a i7 860 or i7 930 vs AMD hex. i3 hyperthreading does help in many apps to speed things along. So does i7 when comparing it to i5 750 in these apps at the same clock speed.

 

[richierich1212]

AzN, you fail to see how most of the top processors in that benchmark all have around the same average fps.

 

[AzN]

Handpick? What are you going on about? Those were Anandtech's benchmarks. These are Anandtech's forums. AzN said i3 was better in games . . . and, at least according to the benchmarks carried out by the people that host these forums, he was wrong.

Anandtech is not god of all hardware review sites. There are plenty of sites that compare i3, AMD x4, and phenom x4 that show i3 beating more games than not. At stock i3 does lose in more games but at overclocked levels on both ends i3 beats anything AMD has to offer.

 

[richierich1212]

Yeah, I guess gaming is all that computers are used for, according to your argument.

 

[Cogman]

Anandtech is not god of all hardware review sites. There are plenty of sites that compare i3, AMD x4, and phenom x4 that show i3 beating more games than not. At stock i3 does lose in more games but at overclocked levels on both ends i3 beats anything AMD has to offer.

Comparing overclocked speeds to just about anything is wrong on many levels. Who's to say the reviewer didn't get a gem or a dude for either chip as far as overclocking goes?

 

[RussianSensation]

So Clarkdale is going to deliver better minimum framerates . . . why exactly?

Unfortunately most games are still not fully multi-threaded. As a result a lot of games are sufficiently maxed out with only 2 cores threads, Clarkdale has 2 threads + 2 virtual threads on top of it giving it enough multi-threading power for most games. Then you are really looking at comparing Phenom II architecture vs. Core i7 architecture per clock cycle. So imagine comparing a Core i7 @ 3.2ghz vs. Phenom II @ 3.2ghz? Given the significant advantage that Core i7 has in performance per clock cycle over Phenom II, it would explain exactly why it's getting better minimum framerates in games that are not heavily multithreaded (of course say in GTA4 the Phenom II x4 will likely win).

Besides, why in the heck are we trying to figure out how Bloomfield will fare against Thuban by comparing Clarkdale to Deneb?

I see how you got confused. I don't think we are extrapolating Clarkdale's to Deneb's performance to Bloomfield vs. Thuban. I think the main argument is that hardly many applications other than digital content creation and video/audio processing will benefit from more than 4 cores. As a result, until games that are more heavily multi-threaded arrive, it is the performance per clock which will be more important for games. Thus, if you compare overclocked Clarkdale to a 4 core Phenom II X4, you are going to see that Clarkdale will outperofrm Phenom II in most games. Even if you overclock Phenom II, you still won't beat Core i7. There is a good article which shows this: http://www.legionhardware.com/artic...ossfire_cpu_scaling_performance_part_2,1.html

Again, this is nothing against AMD 6-core processors. In fact, even a Core i7 930 @ 4.0ghz will be faster than a 6-core 12 threaded Core i7 980X (3.33ghz) in games for minimum framerates.

 

[RussianSensation]

Minimum FPS is a particularly bad measurement because any number of system anomalies can happen which would cause the benchmark to report a lower then expect FPS value. It is 1 measurement verses 1000s.

But if you are swapping various CPUs on socket LGA1156, you are keeping all of your other components constant. Legion Hardware clearly shows that with increased cpu clock speed, minimum framerates increase. I also never said minimums are only affected by CPU speed (of course GPU speed affects them as well).

If you are running a benchmark of a game, you aren't going to start FRAPS at the beginning of you loading the game either, or include cut scenes to scew your results. The proper approach would eliminate situations of 0 fps in less than useful real world scenarios you described (such as loading a screen, or a cut scene where no gameplay actually occurs). So you can plot a section of a benchmark demo over a graph and see how minimums affect gameplay. This is exactly what HardOCP does. Average framerates are often unable to capture excessive loads on a videocard and CPU where frames tank and your game chops.

Tell you what, load Dirt 2 gamedemo and run any level you want with 8AA, 4AA and 2AA on your videocard. You will see your avg frames will be hardly affected compared to minimums which would be cut by 50% if not more. Try playing Dirt 2 with 8AA vs. 2AA now and you'll see if you have an insufficient videocard, the 3x 0.1 millisecond instances where frames drop to 30fps will be choppy as hell, but you'll still get great 50+fps averages. In contrast with 2AA, you will still get 60+ fps average but 48 fps minimums! That's what I am saying. Just like loading a videocard affects minimum framerates, loading a cpu also affects minimum framerates. Not considering minimum framerates does not capture AT ALL what I am going to experience playing a game in the real world.

I posted links to a couple of well trusted and known review sites that show that people that really test components don't look at Minimum FPS (one of those sites being anandtech.) for exactly the reasons I've been saying (without refute from benchzowner)

*Trusted* - by who? is there an international committee that recognizes them as trusted? Do they have strong qualifications in their field, recognized worldwide? That's too ambiguous of a statement. Consumers "trusted" Toyota to make quality cars or for Goldman sachs to make legitimate transactions.

Just because most gaming websites don't use minimum framerates in their benchmarks, doesn't make it the right methodology. 50 years ago, many disputed the harmful effects of Mercury and cigarettes. Those were "trusted" sources. Independent thinking doesn't simply accept the status quo as such. Other websites such as Legionhardware, Xbitlabs, HardOCP and Techreport do include minimums in their testing.

Just think about it for yourself, imagine you are a racecar car driver moving at 300km/h over a 10 km straightaway. Over 3x 300 meter sections you would completely lose engine power. Over the 10 kms, those 600 meters of loss of power will hardly affect avg. speed. However, you would be frustrated I would presume? That's 3 instances of minimum framerates of say 15fps, playing a game at constant 60fps. I guess it comes down to then how susceptible a gamer is to choppiness in gameplay.

 

[AzN]

AzN, you fail to see how most of the top processors in that benchmark all have around the same average fps.

Now average frame doesn't matter? Before that minimum frame rates didn't matter. :hmm: Some of you guys really need to make up your mind. :sneaky:

 

[AzN]

Comparing overclocked speeds to just about anything is wrong on many levels. Who's to say the reviewer didn't get a gem or a dude for either chip as far as overclocking goes?

Oh right because AMD CPU's doesn't quite overclock well as Intel CPU's. :hmm:

Whatever it is i3 is just as good as x4 CPU's in gaming performance if not a little faster when it's all tweaked out. i7 860/930 vs thuban will be no different when an i3 with 2 cores are just as good as AMD's x4 CPU except that i7 will have clear advantage in gaming performance due to having only 66% of the cores instead of 50% of the cores like i3.

 

[DrMrLordX]

Because clarkdale paints a picture how Phenom II x6 is going to compare to i7 860 for instance. Again. You rather compare a $110 processor with a $180 processor. 2 intel core vs 4 AMD core CPU's. I understand i3 is slightly slower than Phenom when comparing @ stock but clock for clock it's very close even in quad optimized games but when overclocked on both ends i3 beats it in more games than a phenom x4. If an i3 with 100% less core is competitive with AMD's best at the moment what makes you think AMD's 6 core CPU is going to compare with intel's i7 4 core with only 50% less core?

Okay, I'm going to say this one more time, and then stop.

1). Clarkdale and Bloomfield/Lynnfield have too many architectural differences for there to be any logical extrapolation of performance involving the two. If you want to know how Bloomfield/Lynnfield will perform in games, the numbers are out there RIGHT NOW. The only unknown is the x6 (see point #2). If you want evidence of this, check out your own benchmark where Clarkdale beats Bloomfield. I guess I should surmise from your benchmark that when Thuban squares off against Bloomfield, it's going to have an easier time of it than Deneb has against Clarkdale?

2). Most games don't really utilize 4 cores properly, much less 6. You want to know how Thuban will do in games? Take a look at whatever the estimated turbo speed for Thuban will be at any given model number, and then look for OCed Deneb results at about that speed. That's x6's performance (thanks to turbo . . . ~3 cores will be factory overclocked). Then compare that to i7-920 with turbo enabled, or i5-750, or i7-860, or whatever.

Anandtech isn't a god of tech benches, but it's good enough for me, which is one of the reasons why I'm here. I don't need to go hunting for game benches at 1024x768 to prove my point.

Unfortunately most games are still not fully multi-threaded. As a result a lot of games are sufficiently maxed out with only 2 cores threads, Clarkdale has 2 threads + 2 virtual threads on top of it giving it enough multi-threading power for most games. Then you are really looking at comparing Phenom II architecture vs. Core i7 architecture per clock cycle. So imagine comparing a Core i7 @ 3.2ghz vs. Phenom II @ 3.2ghz? Given the significant advantage that Core i7 has in performance per clock cycle over Phenom II, it would explain exactly why it's getting better minimum framerates in games that are not heavily multithreaded (of course say in GTA4 the Phenom II x4 will likely win).

This probably more belongs in the video forum than this one, but in all honesty, I have yet to see any clear pattern as to what determines minimum fps. It's all related to the specific timedemo/test run and the conditions under which minimum framerates occurred. I suspect that a lot of it has to do with how well the processor is handling driver calls, and that in turn can be related to cache performance depending on the size of the drivers in memory and the specific optimizations.

To further that point, it should be very telling that, sometimes, I think minimum FPS is more vid card related and NB/Uncore related than it is raw core clock related, beyond a certain point.

I see how you got confused. I don't think we are extrapolating Clarkdale's to Deneb's performance to Bloomfield vs. Thuban. I think the main argument is that hardly many applications other than digital content creation and video/audio processing will benefit from more than 4 cores.

Yes, but Clarkdale's uarch is sufficiently different from Lynnfield/Bloomfield that it has no place in a gaming standoff between Thuban and Lynnfield/Bloomfield. See above where azn's linked bench has Clarkdale beating Lynnfield/Bloomfield?

If you want to look at how x6 will perform in games . . . it's going to be the same as Deneb, with higher results because of turbo. After an overclock (presumably with turbo disabled), it should be the same as an overclocked Deneb at the same clockspeed.

This is why I agree with Rubycon that we should be looking at encoding results here.

 

[aigomorla]

wow.. i cant believe u guys thinnk im wrong about ram...

sigh amd ddr3.... as i call it or old ddr3 had a start up voltage of 1.7v

why do i call it amd ddr3?
because all of intel moved on the lower voltage ddr3, while amd can use the older ddr3 with a 1.7 start up voltage..

when the first lucky people were getting i7.. me being one of them... we were fubarded because only a very few select ddr3 would work because of the low start up voltage...

intel wont let u use this ram because of the .5 rule it breaks @ stock.

this is why we started seeing lower start up voltage ddr3 come out as the i7's got popular. and guess what.. the first sticks said.. designed for i7.

u think i dont know ram??
go look at newegg b4 u challange me onn a statement like this..

u'll see ddr3 @ 1.5 1.65 and 1.7v


sigh... i swear u guys it takes 10 seconds to check my posts at newgg or even tiger direct yet u want to straight up and argue with me...

and how often once again have i been wrong about the i7?

 

[BenchZowner]

wow.. i cant believe u guys thinnk im wrong about ram...

sigh amd ddr3.... as i call it or old ddr3 had a start up voltage of 1.7v

why do i call it amd ddr3?
because all of intel moved on the lower voltage ddr3, while amd can use the older ddr3 with a 1.7 start up voltage..

when the first lucky people were getting i7.. me being one of them... we were fubarded because only a very few select ddr3 would work because of the low start up voltage...

intel wont let u use this ram because of the .5 rule it breaks @ stock.

this is why we started seeing lower start up voltage ddr3 come out as the i7's got popular. and guess what.. the first sticks said.. designed for i7.

u think i dont know ram??
go look at newegg b4 u challange me onn a statement like this..

u'll see ddr3 @ 1.5 1.65 and 1.7v


sigh... i swear u guys 10 seconds to check my posts... and how often once again have i been wrong about the i7?

Actually, you are wrong.
Older DDR3 memory sticks ( DIMMs ) work just fine on most X58 and P55 motherboards, booting up just fine ( by the way the motherboards after a clear CMOS, what we can call a "Cold boot" start with 1.5V being applied to the RAM, and some motherboards start up with 1.6V, and it's not a problem at all, you can modify the BIOS to power up the board at any Vdimm [ Memory Voltage ] ).

The Vdimm - Vtt <= 0.5V rule has been kind of debunked by the way.

And the don't go over 1.8V and such warnings are also false & debunked by lots of overclockers including myself ( I have used up to 2.45Vdimm [ with just 1.4Vtt ] on a few Core i7's [ 2* 920's C1, 2* 920's D0, 2* 950's & 1* 975 ).

p.s. Even with just 1.5V most of the DDR3 ICs out there ( old & new ) will boot up just fine unless the manufacturer's SPD programmers suck

 

[aigomorla]

ive tried to boot up from 1.7v sticks.. and others on the forum as well.

the onnly way i could post was with oem's which would go down to 1.5v

but all my older performance ddr3 onnly ended up as ex pensive paperweights..

i know all ddr3 now comes in the lower voltage... however there are still high voltage ddr3.

and u knnow am3 never has compatability issues with any ddr3