This is why I was asking you before... not to put you on the spot... it just did not make sense to me with what I have experienced.

 

I did one 10x run @ 4.9 (pic does not have cpu-z, just temps - only the filename refers to the 4.9) and temps almost identical to the 2nd pic (same bios settings iirc) - 78/88/87/82 - But this is where I felt I was pushing the chip a bit too hard.

 

I guess our drastically different experiences made us question each other! Just out of curiosity, and in an effort to see if there is another explanation for the voltage difference between your chips and those I've worked with, what RAM speed(s) do you run? I've run a range of RAM frequencies with these chips but all have settled in nicely somewhere above 2000MHz, with my own chip being the most voltage hungry and also using the fastest RAM at 2600 C10. Also, do you use your integrated graphics at all say with Lucid Virtu?

 

 

I prefer a propane torch... heat the IHS until a dark cherry red and it pops right off. Then replace with just a snip of Oatey plumbing solder (again heating until a dark cherry red) and you're back in business.

 

Would you have fine enough control with that so as not to heat up the package or chip too much?

RAM is this G.skill 2666 Cas 11 and I used one of the xmp profiles (1.65v). I really wanted the Corsair Dominator 3000 or this Corsair Dominator 2800 but it was not around at the time.

 

TechMax I thought you had the more expensive G.skill... I think newegg was selling an 8Gb set for around $500 and I think was 2800. I see now a G.skill 2800 Cas 11 kit which looks like it?... just $289. This is what you use?

 

As far as using the the int. graphics (like the lucid) no... w/o bothering to test, I felt like that would add more heat to deal with w/o much (if any) graphics gain.

 

 

Would you have fine enough control with that so as not to heat up the package or chip too much?

 

Just keep the flame moving... I find a butane (kitchen) torch not enough... and oxy-acetylene a wee bit too much.

 

(I hope everyone realizes "we jest")

 

-Rob

RAM is this G.skill 2666 Cas 11 and I used one of the xmp profiles (1.65v). I really wanted the Corsair Dominator 3000 or this Corsair Dominator 2800 but it was not around at the time.

 

TechMax I thought you had the more expensive G.skill... I think newegg was selling an 8Gb set for around $500 and I think was 2800. I see now a G.skill 2800 Cas 11 kit which looks like it?... just $289. This is what you use?

 

As far as using the the int. graphics (like the lucid) no... w/o bothering to test, I felt like that would add more heat to deal with w/o much (if any) graphics gain.

 

 

 

 

Just keep the flame moving... I find a butane (kitchen) torch not enough... and oxy-acetylene a wee bit too much.

 

(I hope everyone realizes "we jest")

 

-Rob

 

Wow Rob, I'm jealous. If I had your chip I'd probably be able to run 5.2w/HT on instead if 4.9 w/HT off since I don't hesitate to add voltage up to 1.5

 

I used to run that GSkill 2800 RAM and it was fantastic stuff but my CPU's IMC couldn't scale that high so I sold it back when it was going for $500 and made some money on the deal because I got lucky and found it when it was only $289. Moving down slightly to 2600 C10 has made no discernible performance impact on my machine (again, I couldn't run the 2800 stuff at full speed, maybe 2680).

Just keep the flame moving... I find a butane (kitchen) torch not enough... and oxy-acetylene a wee bit too much.

 

(I hope everyone realizes "we jest")

 

-Rob

 

I once tried soldering the IHS with an arc welder but the results weren't too pleasant.

 

- Sent from my rooted, Verizon Samsung Galaxy Nexus LTE smartphone via Tapatalk because haters gonna hate.

 

 

RAM is this G.skill 2666 Cas 11 and I used one of the xmp profiles (1.65v). I really wanted the Corsair Dominator 3000 or this Corsair Dominator 2800 but it was not around at the time.

 

TechMax I thought you had the more expensive G.skill... I think newegg was selling an 8Gb set for around $500 and I think was 2800. I see now a G.skill 2800 Cas 11 kit which looks like it?... just $289. This is what you use?

 

As far as using the the int. graphics (like the lucid) no... w/o bothering to test, I felt like that would add more heat to deal with w/o much (if any) graphics gain.

 

 

 

 

Just keep the flame moving... I find a butane (kitchen) torch not enough... and oxy-acetylene a wee bit too much.

 

(I hope everyone realizes "we jest")

 

-Rob

 

Rob, what MOBO are you running?

Rob, what MOBO are you running?

 

Hi Vic,

 

ASUS P8Z77 WS (link to newegg)

I thought about delidding my new i7 3770k which is arriving tomorrow with all the other ordered parts to complete a dedicated FS only machine. My i5 2550k rig is going under my 40inch TV in computer room as gaming/media centre.

 

I may have a go at the delidding process in 6 months or so. I'll make sure the chip works ok first and tempt fate when the next gen of i7's arrives this year. Hopefully the next gen might be worth upgrading to and if so and I make a botch job of delidding at least I can buy the new CPU to make up for it.

To be honest, I was about to delid this very morning, I was minutes away from diving in. But then after a quick calculation it was clear that even if it enabled me to increase my overclock from 4.5 to 5, it still wouldn't amount to much in terms of frames per second.

 

200-300 MHz more and 2-3 frames per second just isn't worth the effort and invalidating the warranty for me.

To be honest, I was about to delid this very morning, I was minutes away from diving in. But then after a quick calculation it was clear that even if it enabled me to increase my overclock from 4.5 to 5, it still wouldn't amount to much in terms of frames per second.

 

200-300 MHz more and 2-3 frames per second just isn't worth the effort and invalidating the warranty for me.

 

Now there's some wisdom. In a double blind experiment I would bet users on a 4.5Ghz machine versus a 5Ghz machine would not show a significant difference in terms of subjective evaluation of 'performance'. I believe this is the case because this scale in difference in clock rate will pall in comparison to differences unique to each flight scenario (plane, weather, scenery complexity). If you are in a taxing scenario that gets you 15fps on your 4.5Ghz machine, at 5.0Ghz you're going to some fraction less than proportional scaling, let's say 70%. This gets you a whopping increase to 16.2fps. Factor in a little more added instability perhaps, and who knows it may stutter more anyway!

To be honest Noel, I think deliding is more about bragging rights.

 

And it's fun of course, and that's okay, it's exciting and brings with it a sense of achievement if we can run our CPU's at 5 as opposed to 4.5. I can relate to that, but sometimes, for some of us, common sense must prevail.

 

I would have to agree with you, objectively, I doubt running then sim at 4.5 as opposed to 5 could be definitively shown to offer much in the way of performance. I wouldn’t envisage any kind of miraculous, nonlinear jump in performance. I envisage a 10% increase in performance max, which amounts to very little at low frame rates, where an increase would be desired.

 

For me, an extra 2 0r 3 frames would only be useful if I were on the boundary between playable, and unplayable, and to be honest, if that were the case, I would be horrified that I'd wasted all that money building a top-notch rig.

 

To put it simply, I don't NEED an extra 2-3 frames per second. And I'd be dammed annoyed if my system was so puny that I did.

 

Who knows for the future though, the next must have add-on may be a frame rate eater, in which case I would think again.

<br />And it's fun of course, and that's okay, it's exciting and brings with it a sense of achievement if we can run our CPU's at 5 as opposed to 4.5. I can relate to that, but sometimes, for some of us, common sense must prevail.<br />

 

Absolutely. I have no problem w/ this as a hobby and striving for whatever you can muster, and 10% bump in total performance is significant, just not that much. As you so well stated, you need the bump when frames are in the toilet, and never will 10% move you out of the toilet! The only time I feel the need to chime in is when the 10% (I doubt in the scenario above it's even that) is framed as the Holy Grail (for performance) when in fact it is what it is, maybe 10%. What really has my wondering now is for myself, why bother? Someone here, maybe it was you, mentioned FSX has some restrictions built-in that impair performance no matter what the hardware. I'm beginning to think this must be the case, though it's certainly counterintuitive. As I've commented many time here, I'm on 4.5y/o hardware now and wow, it seems when I hear people w/ high end parts running at 4.5Ghz and up, report flying in conditions in the 'low teens' it's just appalling and leaves wondering, why bother upgrading until this thing dies of old age? I don't ever see 'low teens'! Lowest ever for me is maybe 15 and that is rare. But, I'm mindful to not ask more than I can put up with. I'm taking off in the QW757 from FTX' KJAC in FTX terrain, most sliders full right, mesh full right, etc, 21-22 frames during TO, and shortly thereafter, 28-30. I'm going to guess if I do go ahead and do the Haswell box and aim for 4.5 or high GHz, I should have a very noticeably better experience, but it's difficult to see where beyond being able to hop in the NGX in MOST places, but geez, apparently not all. This as I say is really where I'm at now. I jump in the NGX, but in select settings only. By looking at CPU charts, Haswell (presuming 10% increase clock for clock over IB) at 4.5Ghz will get very close to 200% CPU performance over my Core 2 Quad at 3.72Ghz. That should clearly demonstrate a big bump in FSX performance, but I can guess from reports here it IS NOT going to be anywhere near 200% performance of what I'm seeing now. Maybe 25%? 30%? That there is a very big restriction somewhere outside the realm of CPU performance must therefore be the case.

I was on the fence a few weeks ago whether to go 3770K and de-lid or 2700K. I ended up with SB and not dissapointed in the least by its performance and OC abilities. Almost perfect chip really.

You de-lidding detractors seem to forget the minimum FPS factor, i.e. the worst case scenario. When your FPS get down to 10 at KJFK at night in the NGX and mine are at 13, which of us do you think is going to have the better simming experience? That's tangible, not subjective.

You de-lidding detractors seem to forget the minimum FPS factor, i.e. the worst case scenario. When your FPS get down to 10 at KJFK at night in the NGX and mine are at 13, which of us do you think is going to have the better simming experience? That's tangible, not subjective.

 

To each his own. I'm afraid the worst case scenario by that definition is totally irrelevant from my perspective because I won't put up with even 13. That's wretched! Much prefer to hop in the QW757 or anything but the NGX if I am at KJFK. Or adjust settings accordingly. Won't reduce my GPU settings but I don't have to fortunately. I only have one display running, and normally I leave autogen at Dense unless I'm in something really light like the Super MD80 Pro or the Turbine Duke runs well. Put together a first rate machine w/ water cooling and delidding and sit at 13? I sure as heck hope to see at least 150% performance boost over my Q9650/GTX280 machine. That will be a very nice increase for me but seriously these comments from the heavyweights here have me wondering if I can expect even that.

Just goes to show how pathetically backwards FSX's code is. The machine is capable, the software is not.