PDA

View Full Version : Msi gt663 eta?



theMethod
11-07-2010, 08:41 PM
Any idea when XoticPC would get and review MSI's upcoming GT663?

SmogHog
11-07-2010, 09:32 PM
It was released in my region on the 25th of October so I would hazard a guess that it will be here any day now.

The 2 models here come without an OS installed.

Here what I posted in another thread.

The MSI GT663 was released in my locale today in 2 flavors,the i7 version and the i5 version.They're built on the Intel PM55 chipset.

The i7 version:
Intel i7 740QM
6gb DDR3 3x2gb @1333(12gb Max)
1TB 500gb x 2 SATA 7200rpm(Raid Available)
DVD Multi
396mm x 269mm x 55mm
3.5kg w/battery
15.6" LED Backlit Display
1920 x 1080 @60hz(16:9)
NVIDIA GTX460M w/1.5gb DDR5
Wireless N
Bluetooth 2.0+EDR
4 USB (2 x 3.0,2 x 2.0)
9 cell battery
HD Webcam
1 HDMI
1 e-SATA/USB 2.0
1D-Sub
Express Card
4 in 1 card reader
OS- Free DOS

i5 version:
Intel i5 460M
4gb DDR3 2gb x2 @1066(8gb Max)
500gb @7200rpm(upgradeable to 1TB 500gb x 2)
DVD Multi
396mm x 269mm x 55mm
3.5kg w/battery
15.6" LED Backlit Display
1920 x 1080 @60hz FHD
NVIDIA GTX460M w/1.5gb DDR5
Wireless N
Bluetooth 2.0+EDR
4 USB(2 x 3.0,2 x 2.0)
9 cell battery
HD Webcam
Express Card
D-Sub
4 in 1 card reader
1 HDMI
1 e-SATA/USB 2.0
OS- Free DOS

theMethod
11-08-2010, 07:50 AM
The 460M would work out better than 5870 (of GX660R) for a single fan for both the CPU and GPU. That's what interests me in the GT663.

powerpack
11-08-2010, 09:34 AM
The 460M would work out better than 5870 (of GX660R) for a single fan for both the CPU and GPU. That's what interests me in the GT663.Notebookcheck has the 460 and 5870 on par on power consumption. Is there any other reason you think 460 better on single fan?

coyoteblue
11-08-2010, 12:55 PM
I was under the impression that the 460 ran cooler than the 5870. Is that not the case, or only specific to the Asus G73 series? :confused:

Maverick494
11-08-2010, 12:57 PM
I was under the impression that the 460 ran cooler than the 5870. Is that not the case, or only specific to the Asus G73 series? :confused:

That is only so far shown to be true in the G73 series and it likely has to do with some changes that were made between the JH and JW cooling systems.

I haven't seen any temps on the 8690 or 8760 with the 460M in it yet to compare.

theMethod
11-08-2010, 01:06 PM
Notebookcheck has the 460 and 5870 on par on power consumption. Is there any other reason you think 460 better on single fan?

I don't know if Asus made any design changes in the G73JW internally compared to JH, but the former (with 460) does have significantly better temperatures than the latter (with 5870). Even if I lose a couple of frames every second, a cooler running card when only having the liberty of one fan seems like a good idea.

Also, Nvidias actually tend to idle at low power consumption and save battery, but ATIs hardly make a difference. I cannot find that web page now but when Sager's 8690 first offered 5870 as another alternative to GTX 280, a review showed the battery life for just surfing the web with 5870 was 40 minutes, and with 280 was 55 minutes. Yeah I know it has a pathetic 3-cell 3800mAH battery...

powerpack
11-08-2010, 01:14 PM
Your logic on the 460 makes sense if the ATI runs hotter or uses less batt. Xotic said th JW has improved thermal solution over JH.

Maverick494
11-08-2010, 01:14 PM
I don't know if Asus made any design changes in the G73JW internally compared to JH, but the former (with 460) does have significantly better temperatures than the latter (with 5870). Even if I lose a couple of frames every second, a cooler running card when only having the liberty of one fan seems like a good idea.

Also, Nvidias actually tend to idle at low power consumption and save battery, but ATIs hardly make a difference. I cannot find that web page now but when Sager's 8690 first offered 5870 as another alternative to GTX 280, a review showed the battery life for just surfing the web with 5870 was 40 minutes, and with 280 was 55 minutes. Yeah I know it has a pathetic 3-cell 3800mAH battery...

Funny, that isn't what I got on battery. I got 1 hour 47 minutes on battery with my 5870, so perhaps you misread that.

Plus the GTX285M isn't even comparable to the 5870 in any way.

You're also going to lose a lot more than 1-2 FPS, the average is about 7-10. Running cooler only means anything if A) You are trying to overclock or B) you are running higher than the TDP of the card. Since B isn't true even on the G73JH (except in some very rare cases) then A is the only other reason.

Single fan solutions can work just fine, do you know how many CFM the dual fans of the Sager or G73JH are pushing? Do you know how many CFM the single fan of the MSI is pushing? if you answer no to either or both of those questions you can't really make a good comparison.

If, for instance, MSI is using a much larger single fan that is pushing 50 CFM peak load and Sager/Asus are using two smaller fans that are pushing 25 CFM peak loak EACH, then (assuming airflow is evenly distributed) the MSI should have the same potential cooling. The key is if MSI distributed the airflow well and ensured that their heat sinks are efficient then there really should be no problem.

theMethod
11-09-2010, 05:18 AM
Funny, that isn't what I got on battery. I got 1 hour 47 minutes on battery with my 5870, so perhaps you misread that.

Well, if you really have to exaggerate, then please do so up to some believable limits. It's not possible for a battery supplying just 42 watts every hour to support a 5870, i7 with four cores and a 7200rpm HDD for that long.

Check out
http://www.xoticpc.com/sager-np8690-built-clevo-w860cu-custom-gaming-laptop-pri-2659.html?reviews_id=123
He says "At best I get 1 hour from the battery".
and
http://www.notebookreview.com/default.asp?newsID=5406&p=2
Scroll down to the battery life section where they say "only 57 minutes of battery life while surfing the Internet with the screen brightness one notch above minimum".


Plus the GTX285M isn't even comparable to the 5870 in any way.

Agreed (with an "obviously" shrug), but quite irrelevant to the discussion, especially knowing that where the 460M has a TDP of 72 watts, the 285M has 75 and does not have the Optimus feature, stealing another couple of minutes from the overall battery life while using non-GPU-intensive applications.


Running cooler only means anything if A) You are trying to overclock or B) you are running higher than the TDP of the card. Since B isn't true even on the G73JH (except in some very rare cases) then A is the only other reason.

I assume you meant "Running hotter only means..."
Well, we both know the JW does turn up better temperatures than the JH, and that's with no overclocking on either.


Xotic said th JW has improved thermal solution over JH.

I read up a bit and that's what a couple of reviewers (including XoticPC) are saying, although nowhere do they really tell exactly what improvements were made, as in, if the fans were increased in size or if they spin faster or the timing of their spinning up, etc. The insides of both the JH and JW appear similar in the videos.


do you know how many CFM the dual fans of the Sager or G73JH are pushing? Do you know how many CFM the single fan of the MSI is pushing? if you answer no to either or both of those questions you can't really make a good comparison.

Knowing that would really make everything so transparent! I doubt it really gets out beyond Clevo's, Asus's or MSI's labs, unless someone would be kind enough to test them out and put up how much volume of air do their fans push. I only know that logically, unless each of the two fans (of either the G73 or the 8690) is less than 2/3rd the size and spins at less than 2/3rd the speed of GX660R's fan, they push (in sum) more air than the latter.

Maverick494
11-09-2010, 12:17 PM
Well, if you really have to exaggerate, then please do so up to some believable limits. It's not possible for a battery supplying just 42 watts every hour to support a 5870, i7 with four cores and a 7200rpm HDD for that long.

Check out
http://www.xoticpc.com/sager-np8690-built-clevo-w860cu-custom-gaming-laptop-pri-2659.html?reviews_id=123
He says "At best I get 1 hour from the battery".
and
http://www.notebookreview.com/default.asp?newsID=5406&p=2
Scroll down to the battery life section where they say "only 57 minutes of battery life while surfing the Internet with the screen brightness one notch above minimum".

so you just called me a liar? hey thanks johnny-come-lately. Here is my take on it: They weren't getting the video card to properly downclock. I had to downclock it on my own.

Thanks for showing you know exactly dick about a laptop you don't have.




Knowing that would really make everything so transparent! I doubt it really gets out beyond Clevo's, Asus's or MSI's labs, unless someone would be kind enough to test them out and put up how much volume of air do their fans push. I only know that logically, unless each of the two fans (of either the G73 or the 8690) is less than 2/3rd the size and spins at less than 2/3rd the speed of GX660R's fan, they push (in sum) more air than the latter.

I'd like to see the math where you pulled a 67% size difference out of your ass from.

Vertrucio
11-09-2010, 02:09 PM
Uh, ease off the hostility there man. No need for that here.

Maverick494
11-09-2010, 02:19 PM
Uh, ease off the hostility there man. No need for that here.

there is also no need for someone who doesn't even own the laptop to flat out call someone who does a liar.

theMethod
11-09-2010, 09:47 PM
I'd like to see the math where you pulled a 67% size difference out of your ass from.

Calculating the absolute value of how much air a fan pushes in every revolution would require the knowledge of all aspects of not only the fan itself (like: number of blades, tilt angle of blades, length and width of each blade, etc.), but also the external factors (like: the total area blocked by the grill on the fan, the volumetric capacity of the area to be cooled, the number of times every minute the entire volume of air in there has to be completely displaced, etc.). However, if comparison is made, we have a straightforward formula for the speed to be considered exclusively, considering the other factors of the two fans being compared would be equal:
CFM2 = CFM1 * ( RPM2 / RPM1 )

Assuming Fan1 is rated at 50 CFM and spins at 100 RPM, then the CFM of Fan2, which spins at 67% the rate of Fan1 ((2/3)*100 = 66 RPM approx.) would be:
CFM2 = 50 * ( 66 / 100 )
CFM2 = 33 (approx.), that is 2/3rd of CFM1 (50).

So, if other factors were equal, then 2 fans of type Fan2 would push a sum of 66 CFM of air, compared to 1 fan of type Fan1, which pushes 50 CFM; that is, approx. 4/3rd more air.

Now, we have CFM = Pi * (fan radius)^2 * Column length of air pushed

Where "column length of air pushed" means:
If each revolution of the fan pushes 0.5 feet of air at 100 RPM, then it pushes (100 * 0.5 =) 50 feet of air in 1 minute. It can also be seen as a 50 feet long column of air with a diameter equal to the diameter of fan being pushed every minute.

As I said above, acquiring the knowledge of all factors related to finding the exact amount of air being pushed every revolution is beyond my capability. So, we'll assume a value for fan and make a comparison again, but this time with respect to size.

Assuming Fan1 has a radius of 0.3 feet and the column of air it pushes every minute has a length of 10 feet, its CFM would be:
CFM1 = 3.14 * (0.3)^2 * 50
CFM1 = 14.13

If the size of Fan2 is 67%, that is 0.2 feet, and assuming that due to some other factors being better, it pushes the same column length of air every minute, then:
CFM2 = 3.14 * (0.2)^2 * 50
CFM2 = 6.28, which is 2.25 times less than (or 4/9th of) CFM1.

So, two fans of type Fan2 here would have a CFM of (2 * 4/9) 8/9th that of 1 fan of type Fan1. From first result, increase is (4/3 =) 1.33 - 1 = 0.33 times more and decrease is (8/9 =) 0.88 - 1 = -0.22 times less, with the assumption of an equal length of column being pushed. In reality, the length of column would be significantly lesser and the decrease would therefore be greater, approximately equalling -0.33 times; which is the increase exactly. So in total, +0.33 + (-0.33) = 0.

Looks like even my ass is better at math than your ego-stuffed brains.



so you just called me a liar? hey thanks johnny-come-lately. Here is my take on it: They weren't getting the video card to properly downclock. I had to downclock it on my own. Thanks for showing you know exactly dick about a laptop you don't have.

If you wish to be so direct in your communication, then yes, I did call you a liar. I'll let the figures do the talking themselves, a battery supplying 42 watts an hour trying to keep running a 5870, i7 with four cores and a 7200 rpm HDD. No matter how much you downclock your GPU, you'll never squeeze out more than 70 minutes, unless you simply disable its usage. You're only making a fool out of yourself here with this attitude.



Uh, ease off the hostility there man. No need for that here.

I appreciate it Vertrucio, but it's quite alright with me. I cannot blame him because I was almost the same when I started using online forums a couple of years ago, always used to put my ego before myself. I expect him to be more mature in an year.

Maverick494
11-09-2010, 10:54 PM
Calculating the absolute value of how much air a fan pushes in every revolution would require the knowledge of all aspects of not only the fan itself (like: number of blades, tilt angle of blades, length and width of each blade, etc.), but also the external factors (like: the total area blocked by the grill on the fan, the volumetric capacity of the area to be cooled, the number of times every minute the entire volume of air in there has to be completely displaced, etc.). However, if comparison is made, we have a straightforward formula for the speed to be considered exclusively, considering the other factors of the two fans being compared would be equal:
CFM2 = CFM1 * ( RPM2 / RPM1 )

Assuming Fan1 is rated at 50 CFM and spins at 100 RPM, then the CFM of Fan2, which spins at 67% the rate of Fan1 ((2/3)*100 = 66 RPM approx.) would be:
CFM2 = 50 * ( 66 / 100 )
CFM2 = 33 (approx.), that is 2/3rd of CFM1 (50).

So, if other factors were equal, then 2 fans of type Fan2 would push a sum of 66 CFM of air, compared to 1 fan of type Fan1, which pushes 50 CFM; that is, approx. 4/3rd more air.

Now, we have CFM = Pi * (fan radius)^2 * Column length of air pushed

Where "column length of air pushed" means:
If each revolution of the fan pushes 0.5 feet of air at 100 RPM, then it pushes (100 * 0.5 =) 50 feet of air in 1 minute. It can also be seen as a 50 feet long column of air with a diameter equal to the diameter of fan being pushed every minute.

As I said above, acquiring the knowledge of all factors related to finding the exact amount of air being pushed every revolution is beyond my capability. So, we'll assume a value for fan and make a comparison again, but this time with respect to size.

Assuming Fan1 has a radius of 0.3 feet and the column of air it pushes every minute has a length of 10 feet, its CFM would be:
CFM1 = 3.14 * (0.3)^2 * 50
CFM1 = 14.13

If the size of Fan2 is 67%, that is 0.2 feet, and assuming that due to some other factors being better, it pushes the same column length of air every minute, then:
CFM2 = 3.14 * (0.2)^2 * 50
CFM2 = 6.28, which is 2.25 times less than (or 4/9th of) CFM1.

So, two fans of type Fan2 here would have a CFM of (2 * 4/9) 8/9th that of 1 fan of type Fan1. From first result, increase is (4/3 =) 1.33 - 1 = 0.33 times more and decrease is (8/9 =) 0.88 - 1 = -0.22 times less, with the assumption of an equal length of column being pushed. In reality, the length of column would be significantly lesser and the decrease would therefore be greater, approximately equalling -0.33 times; which is the increase exactly. So in total, +0.33 + (-0.33) = 0.

Looks like even my ass is better at math than your ego-stuffed brains.

except you just made an ASS ton of assumptions: Like blade pitches are exactly the same and rotational speeds at peak loading. Which means that your numbers are based completely on speculation with no basis in fact. So it looks like you just pulled a bunch of numbers out of your ass to try and make your point. Your math is technically accurate, the problem is you base it on numbers that aren't even known.

You stated that you can't see how a single fan setup would cool as well as a dual fan setup. You just proved exactly how any engineer would do it. Find out the numbers on the smaller dual fan setup and then make your # of blades, pitch and size equate as closely as possible with those from the dual fan setup. Or do you believe that the engineers that work at MSI aren't capable on that math?




If you wish to be so direct in your communication, then yes, I did call you a liar. I'll let the figures do the talking themselves, a battery supplying 42 watts an hour trying to keep running a 5870, i7 with four cores and a 7200 rpm HDD. No matter how much you downclock your GPU, you'll never squeeze out more than 70 minutes, unless you simply disable its usage. You're only making a fool out of yourself here with this attitude.

Except I own the laptop, I sat there with it at about 15% brightness, BT on, Wireless on, and I surfed the web and checked email. I got 1:43 minutes from full charge to 5% battery left.

the 2.5" hard drive uses about 1.5 W(http://www.wdc.com/en/products/products.asp?driveid=854), the i7 at low load is about 13W(http://download.intel.com/design/processor/datashts/320765.pdf) and the 5870 is about 10W at low load (downclocked to 250/400 @.95V) this is the only number I am estimating since it is effing impossible to find an actual datasheet on this damn thing.

that means you are using about 24.5W in low power mode downclocked.

Assuming 42.14 Wh that means you get 1.72041 hours out of the battery. So best case you get 1.75 hours out of the battery. If you juice up the GPU and make it run videos or something you cut battey life a bit, maybe even to your 70 minute mark, but if all your doing is some web surfing and light email, 1:45 minutes isn't impossible.

next time run your mouth to someone that doesn't own the laptop and has tested it.

theMethod
11-10-2010, 01:30 PM
except you just made an ASS ton of assumptions: Like blade pitches are exactly the same and rotational speeds at peak loading. Which means that your numbers are based completely on speculation with no basis in fact. So it looks like you just pulled a bunch of numbers out of your ass to try and make your point. Your math is technically accurate, the problem is you base it on numbers that aren't even known.

To remind you, all I was doing was showing you how I ended up with the 67% size and speed mark. All the rest of the factors, I obviously made an assumption to be somewhat equal in both the fans. I can't the see the reason you are complaining now, after getting the detailed math you asked for.


You stated that you can't see how a single fan setup would cool as well as a dual fan setup. You just proved exactly how any engineer would do it. Find out the numbers on the smaller dual fan setup and then make your # of blades, pitch and size equate as closely as possible with those from the dual fan setup. Or do you believe that the engineers that work at MSI aren't capable on that math?

No, I certainly believe that engineers in MSI are more than capable of designing adequately powerful cooling systems for their laptops. However, engineers everywhere are always stopped short from completely implementing their knowledge by the money-saving business heads of the companies. It actually might be that this single fan in GX660R cools better than two fans of either 8690 or G73, but considering the fact that we don't know factors needed to calculate their CFM, what would you expect the chances are of that being true? Does it really sound very reasonable that a single fan would cool both a CPU + GPU more effectively than a two-fan set-up (one fan dedicated to each the CPU and GPU)?


Except I own the laptop, I sat there with it at about 15% brightness, BT on, Wireless on, and I surfed the web and checked email. I got 1:43 minutes from full charge to 5% battery left.

the 2.5" hard drive uses about 1.5 W(http://www.wdc.com/en/products/products.asp?driveid=854), the i7 at low load is about 13W(http://download.intel.com/design/processor/datashts/320765.pdf) and the 5870 is about 10W at low load (downclocked to 250/400 @.95V) this is the only number I am estimating since it is effing impossible to find an actual datasheet on this damn thing.

that means you are using about 24.5W in low power mode downclocked.

Assuming 42.14 Wh that means you get 1.72041 hours out of the battery. So best case you get 1.75 hours out of the battery. If you juice up the GPU and make it run videos or something you cut battey life a bit, maybe even to your 70 minute mark, but if all your doing is some web surfing and light email, 1:45 minutes isn't impossible.

next time run your mouth to someone that doesn't own the laptop and has tested it.

Well, does your 8690 use solar power to spin the two fans inside it? It's more than just those three components that will eat up your battery, I only quoted those three because they each are a step ahead of what is standard on most laptops, and would consume more power. If all components are considered, standard or not, then most of the power is actually consumed by the cooling fan (or fans, in your case). Even when it's almost off, a single fan with be consuming 20W! (http://www.codinghorror.com/blog/2008/04/revisiting-how-much-power-does-my-laptop-really-use.html) Assuming your GPU fan is consuming 20W and CPU fan 21W (using the data from the above source) when you are just surfing the web, your total power consumption now becomes 24.5 + 41 = 65.5W! I'll re-post these links I posted earlier:
http://www.xoticpc.com/sager-np8690-built-clevo-w860cu-custom-gaming-laptop-pri-2659.html?reviews_id=123
He says "At best I get 1 hour from the battery".
and
http://www.notebookreview.com/default.asp?newsID=5406&p=2
Scroll down to the battery life section where they say "only 57 minutes of battery life while surfing the Internet with the screen brightness one notch above minimum".

Maverick494
11-10-2010, 03:25 PM
Well, does your 8690 use solar power to spin the two fans inside it? It's more than just those three components that will eat up your battery, I only quoted those three because they each are a step ahead of what is standard on most laptops, and would consume more power. If all components are considered, standard or not, then most of the power is actually consumed by the cooling fan (or fans, in your case). Even when it's almost off, a single fan with be consuming 20W! (http://www.codinghorror.com/blog/2008/04/revisiting-how-much-power-does-my-laptop-really-use.html) Assuming your GPU fan is consuming 20W and CPU fan 21W (using the data from the above source) when you are just surfing the web, your total power consumption now becomes 24.5 + 41 = 65.5W! I'll re-post these links I posted earlier:
http://www.xoticpc.com/sager-np8690-built-clevo-w860cu-custom-gaming-laptop-pri-2659.html?reviews_id=123
He says "At best I get 1 hour from the battery".
and
http://www.notebookreview.com/default.asp?newsID=5406&p=2
Scroll down to the battery life section where they say "only 57 minutes of battery life while surfing the Internet with the screen brightness one notch above minimum".


Get back to me when you own the laptop and sit there and run the battery test yourself. Until then you don't know anything about what you are talking about.

I ran the test, I know that the results were.

FTR the fans in low power mode are very very slow. So no they aren't using 20W of power.

You might want to get out a source that isn't 2 years old and a blog.

Maverick494
11-10-2010, 06:25 PM
Just to reinterate the fact that the blog you link to is ver clearly wrong:

http://www.sparepartswarehouse.com/Toshiba,Laptop,Part,P000412700.aspx

5V @ 200mA = 1W (that's peak load)

so some how that guys fans are running 20x more power (4 AMPS) than a normal laptop fan.

So even if your fans were 2x that much power draw you are looking at 2W for the fans, not 20.

so what do we know:

Fans: 1.5W
HDD: 1.5W
CPU: 13W
GPU: 10W (est. downclocked)
LCD: 7W at full brightness so at 15% let's say its at 4W(http://lcdtech.no-ip.info/en/data/laptop.lcd.panels.htm)
Let's add 2W for incidentals like wifi and bluetooth (which is way generous)

total: 32W

so worst case (I estimated more than likely way high) you are looking at about 1.32 hours.

Maverick494
11-10-2010, 09:07 PM
since I posted another post the is sitting in lala land waiting on mod approval I will repost it without the links and add the links later:

LCD: 7W @ max bright. Let's assume 4W at minimum(which is probably damned high). (http://www.yrccjc.com/showroom/hs5e3omqs/p66143/CPT-15-4-quot-TFT-LCD-panel.html shows 6.3W max, I allowed some leeway for the better FHD screen)
CPU: 13W
HDD: 1.5W
Fans: 1.5W (5V @300mA) (http://www.sparepartswarehouse.com/Toshiba,Laptop,Part,P000412700.aspx)
GPU: 10W (@.95V 250/400 clocks, est.)
Extra just in case cushion: 2W

So worst case you are looking at 35W or 1.2 hours. That is assuming I am not way high on the LCD and the cushion, which I bet I am.

You're flat out wrong. There is no way in hell a laptop fan is using 20W. It would have to be 4A draw which desktop fans don't even draw.

Maverick494
11-10-2010, 09:24 PM
and further reading for those that think I am full of it. This is a paper done that has a much older laptop (therefore less efficient) but it illustrates the power draws quite well and they are very close to my numbers:

http://users.crhc.illinois.edu/mahesri/classes/project_report_cs497yyz.pdf

wireless is going to draw about the same, LCD is probably going to draw more and according to that paper his drew 3.5W full bright and I allow for more than his 3.5W full bright on a full dimmed LCD. CPU draws I have proven, along with HDD draws. Fan draws I have also proven to be far lower than your claim. I even added cushion and assuming that worst case scenario is that you have for some reason drawn more power than I anticipated. There is a very likely possibility of getting 1.5 to 1.75 hours of life out of a fresh battery if you really go all out on conservation.