DFI LT X48-T2R Review

kitfit1 said:
I take it then you did'nt bother to check my OCDB link
Click to expand...
Yes I did

kitfit1 said:
I did'nt say that cpu speed had anything to do with GTL settings.
Click to expand...
kitfit1 said:
...If you click the link in my sig you will find the GTL's needed to run at 4.6ghz are completely different from that list...
Click to expand...
:confused:

kitfit1 said:
In a round about way i was saying that the values in your chart simply will not work. When you get your phase unit set up, you will see exactly what i mean. Try using that chart to set GTL values will result in instant BSODS as soon as you run Orthos/Prime. I know because i have a stable system with my values, and imputing yours caused an immediate BSOD in Orhtos.
Click to expand...

Well, what can I say... the values in the chart were only given to me by the Head DFI Engineer...

They do exactly the same as the x?? settings on an Asus board except they allow for a little more adjustment... Plus looking at your link again with the exception of your NB GTL's you have them virtually set at the numbers in the 61% chart anyway, so I dont really understand what you're getting at tbh :confused:

Plus, why or earth would anyone want or need to run Orthos... total waste of time IMO... It proves nothing, it's not a real world test! It stresses the CPU to basically 100%, which no game & or program ever does & it doesnt take other components into account in the mean time... Stability is just that, I dont bother with prime anymore & have never EVER had an issue with crashes BSOD's or anything of the like... The last time I ran orthos was with my Q6600 & it told me that my CPU wasnt stable @ 3.8Ghz with 1.5vcore... Funny how it ran anything & everything for nearly 9months without a single crash or hiccup...

BALLS to PRIME :D
 
Last edited:
With modern (and better) operating systems more sophisticated algorithms were developed to recover from fatal error. However, if a system is not cabaple of running programs like orthos "indefinately" without corruption is not considered stable. Not stable "scientifically" at least.

For everyday use and general computing on the other hand you will not notice either because the operating system recovers from it, or the hardware is not pushed to the limit.
 
drak3 said:
With modern (and better) operating systems more sophisticated algorithms were developed to recover from fatal error. However, if a system is not cabaple of running programs like orthos "indefinately" without corruption is not considered stable. Not stable "scientifically" at least.

For everyday use and general computing on the other hand you will not notice either because the operating system recovers from it, or the hardware is not pushed to the limit.
Click to expand...

Yea, but it's still BALLS & a waste of Leccy IMO...

This is how my stability test regime works...

1. It Posts...
2. It boots into Windows...
3. Runs All Benchmarks 32M Pi, all 3dMarks & PCmark
4. Renders a Full Screen in Cinebench

I usually do that, then downclock 100mhz & undervolt a little, hey presto... Stable system. I dont need some idiot program to tell me that my PC is stable. :p

The only time I ever use it is if I am having problems finding a good clock on a new CPU that wont do what it should... I had issues with my first E8500 that happened to be a bit of a bobbins chip, it passed 10hrs prime only to randomly reboot when starting Firefox or watching a film. :(
 
I'll tell you what Monstermunch, you believe what you want to believe and i'll do the same. The difference is of course, i'm running 4.8ghz stable 24/7 and you're not. A 4.8ghz CPUZ suicide shot proves nothing. Out of interest, why remove the watermarks from you pictures ?
 
You guys might also like these links
DFI LT X48-T2R Living Review GTL tables compiled with the input from immy & Jarry @ DFI.
It's an ongoing review latest addition is the DFI X38/X48 Voltage Measuring Points.
also the
DFI Lanparty P35/X38/X48 BIOS Setting Guidelines

At least Clunk's work is all original :D
r1ch - owner of Digital Report
Hi Clunk, I'm the owner of Digital Report.

I've replied to your email, but I just wanted to publicly apologise for the use of your images - Mike sent them all to me and I added the watermarks believing they were his own. They've already been removed/replaced.

Hope the rest of you liked the review. Let us know if there's anything you'd like to see in future reviews
Click to expand...

luck:)
 
Last edited:
supershanks said:
At least Clunk's work is all original :D
Click to expand...

You're an idiot... :D

The only thing that I apparently had from Clunk, was a LanParty logo that looked like it was made by DFI themselves & probably technically the property of DFI in anycase because of their trademark of the LanParty logo...:confused:

I apparently had a picture that was from clunk's site, which was news to me as I found it by googling the word "HeatMinator"... The Picture was just of the heatsink standing on it's own... so who cares, what difference does it make :confused:

This whole image stealing thing it's utter bull**** to be honest, there is no copyright on the images, I wasn't deeplinking, plus I had permission from DFI to do the Review.
 
Last edited:
A good review Monstermunch. The point raised about documentation for some of these 'advanced' features is a valid one. Monstermunch mentioned to DFI that some written documentation wouldn't go a miss. Personally I don't think DFI need to explain these features as they can get very techie and only a minority would benifit from these tweaks when going for extreme overclocking.

kitfit1 has a point about using slightly different settings in the bios for phase, I would hazzard a guess that kitfit1 and monstermunch are using the same bios? After all, settings from the stock bios may not work with the latest bios, as Monstermunch found out ;)
 
The Q9550 was stable at about 3.7Ghz IIRC. Which would equate to a Q6600 at anywhere between 3.9 and 4.2ghz depending on the task.

/OT

:)
 
weescott said:
The Q9550 was stable at about 3.7Ghz IIRC. Which would equate to a Q6600 at anywhere between 3.9 and 4.2ghz depending on the task.

/OT

:)
Click to expand...

Interesting

How is DFI with corsair? Any problems? I decided on a corsair 2x2GB kit and now considering the DFI mobo and an Q6600 or a new 45nm quad.
 
Bikerz said:
I bet it is the PVM's but im not saying it is and you blaim me if it isnt. What temp do yours get upto? (System temp on smart gardian) Tho ppriming should max out *** PVM's more than gaming so maybe not. Maybe more NB voltage, I had to up my northbridge now I have a 8800 GTX in, seeing as GFX bandwidth goes through NB it needed a bit more juice, which would explian why it crashes only when gaming
Click to expand...

was playing call of duty just now, lasted for 3 maps, then crash. PWMs got to around 46c. That was check in bios whilst restarting pc from crash so probs a bit higher. At what temps to they start to become unstable?

WJA96 said:
Where I have had issues with systems passing prime then failing when playing games, I tend to look at the PSU as often the graphics draws huge power from the PSU in a way that doesn't happen when it's priming. Does it run 3DMark '06 OK? What is the PSU? I managed to blow up a Tagan 700W with a pair of X1950 Pro's in CrossFireX with mine, so I reckon it needs a hefty PSU with a big graphics card(s). I would also check the automatic voltage adjuster - I have mine set to 100.23%, but I have seen it add big chunks of voltage when it's set on AUTO. It may be it's fast enough to pick up the load on CPU/RAM stress testing, but not on games.
Click to expand...

Im think it may be the psu you aswell, though it did run 3dmark and Crysis benchmark fine. My 24pin power connector is a little bit iffy though as in if I knock it the pc will lose power and reboot.

My voltage adjust is currently set at AUTO so will give 100.23% a try now.
 
SpudMaster said:
was playing call of duty just now, lasted for 3 maps, then crash. PWMs got to around 46c. That was check in bios whilst restarting pc from crash so probs a bit higher. At what temps to they start to become unstable?
Click to expand...

DFI gives a max safe temp for the 8 phase digital PWM's as 40c. As you start to go over that temp the PWM losses it's ability to regulate the power properly.
 
WJA96 said:
That could be it then:D
Click to expand...
yeah i know, im just a little reluctant to spend £100 on a new psu n find out it wasnt the problem. Current psu is an enermax noistaker 600w. Had it for 2 - 3 years so maybe its getting tired?

kitfit1 said:
DFI gives a max safe temp for the 8 phase digital PWM's as 40c. As you start to go over that temp the PWM losses it's ability to regulate the power properly.
Click to expand...

40c seems a little low ?
 
How am I an idiot quoting facts ??
These are the pictures that were used :-
lanparty-logo3.png

&
IMG_2070.jpg


It may be that your eyes are bad & you missed the watermarks. If so I apologise.
:D

But if you have the board it's surely easier & quicker to take a picture of your own kit than
I found it by googling the word "HeatMinator".
Click to expand...
:lol:


luck:)
 
SpudMaster said:
40c seems a little low ?
Click to expand...

I know it does. But if you go to DFI Club and do a search for Digital PWM temps, you will find that is what DFI say. It's not that difficult to keep them well below 40c anyway. I have 2 40mm fans glued to the PWM heatsink and even at full dual Prime load it never goes above 38c.
 
kitfit1 said:
I know it does. But if you go to DFI Club and do a search for Digital PWM temps, you will find that is what DFI say. It's not that difficult to keep them well below 40c anyway. I have 2 40mm fans glued to the PWM heatsink and even at full dual Prime load it never goes above 38c.
Click to expand...

at dual prime load im stable for 25 n half hours, no errors. Just put a 60mm akasa fan blowing over the pwm area. Played COD4, 5 mins and locks up :mad:. Upon rebooting i check temp in bios and was at 41c.

It crashes at my current overclock, even though prime stable and when at stock :mad:
 
You must log in or register to reply here.
Back
Top Bottom