G-Omaha's **8790 Review** Now Open To Questions and Comments - Page 7

Sorry to hear about your issues G-Omaha. You have alot more patience than I do. I would have sent it back already.

Good Luck, hope it works out.

Well, I think I mentioned that in the very first thread here in the forum about the 8790... since I don't know where it exactly is (and I'am too lazy now to search after it) I will sum up here my impressions from that review day out of my memory.

I reviewed the new 8790/D870P quickly some weeks ago when I picked up my 8890 from my domestic european dealer (my domestic dealer was one of first who had the Clevo D870P in europe...). During the three hours at my dealers place, where I've setup the initial installment of my 8890, I tested and played also around with the 8790 in parallel. I didn't had MM handy or installed on the 8790, so I can't tell you exact temperature/heat measurements. First the 8790 didn't had any installed OS and so I've fired up an bootable Linux based CD (similar to KNOPPIX) inside the 8790. The 8790 Screen was better than I had expected (no dead or stuck pixels etc.), build quality and the feeled heat were in normal parameters. The keyboard got slightly warm as the one side and bottom of the 8790. - So far so good and nothing unspectacular odd...

Later we (my dealer and I) also installed XP Pro on the 8790 since I was/had already doing the same to my new 8890 at his side. During installing the XP Pro on the 8790, I went to another desk were my 8890 was and installed a wireless miniUSB card into my 8890 and afterwards installed the accompanied wifi drivers in order to test the wifi card. Now the 8790 began somehow to heat up and it's fans kicked in, so that I couldn't hear my 8890 anymore (I was ~three meters away from the 8790). The fan sounding of the 8790 when on full speed state, were like a high frequency airplane turbine. I two times went to my dealers kitchen, which is some rooms away from the place were we installed the machines and I still heared the loud fans of the 8790 when they kicked in like a turbine, even the 8790 were not stress tested at all, just performing on it plain Windows installments. - The 8790 was configured like this, a P4-C 3.4GHz Nothwood CPU with a 80 gig hdd.

My 8790 sum up/verdict:

The 8790 (Clevo D870P) is a pretty good looking and sturdy machine, with a good display and keyboard. I would call it's overall build quality sturdy good and even it's default coloring is really fine. - The only bad points on this machine are ...no need to ask... it's actual high degree fan noise. Also as your individual unit shows, some units may suffer from higher heat problems too.

A word about the 8790 heatpipes:

AFAIK, the assembling concept of the 8790 heatpipes has been made more foolprove by Sager/Clevo. This means the 8790 heatpipes have a plate with a protective layer underwear. When such a heatpipe is assembled one has just to remove that protective layer of the heatpipe plate and then to place the heatpipe into the notebook. This heatsink can't usually be wrong assembled, since it only fits in one direction into the notebook. - If somebody assembles the heatpipe and forgets to remove the protective strip layer, the notebook would overheat and should usually shut down itself.


A word about the fans:

The 8790 fans are special build ones for the notebook and some people are not this happy with these fans, since they tend to make high frequancy noise when they are rotated with much speed. AFAI have heard, Clevo already knows this and is looking to improve these for future series models in some manner. Maybe they can lower noise with some BIOS updates, or they will replace the fans with others for future models...who knows...and time will show...


Don't know if the above is of any help for you, but maybe it points you into the right direction...(?)

Did you tried if CPU-Z maybe can list the specs and manufactor of your memory correctly via it's memory tab page?

See: http://www.cpuid.com/cpuz.php

@ G-Omaha

Ok dude, I one hour ago pleased my Clevo buddy to do some forced heat stress-testing for the D870P (8790). We wanted to see if we are able to reproduce the same behaviour/problems as you are encountering with your machine (BTW, we did the test via a telephone conferencing session and thus I could hear the soundings of the machine slightly too...). The 8790 used now for this test had a P4-C 3.4 GHz CPU assembled und two huge 80 GB disks inside running under RAID level 0.

My buddy fired up some huge stress testing app and closed the bottom fans of the 8790 completely with a cushion. The machine needed a very long period (we were waiting, waiting and waiting...) since it slowly heated up, but finally he/we could force it up until it reached 69°C degrees. He said that the cooling fans got some faster speed up (e.g. they rotated faster as usual...) but the noise hasn't got much louder as if the machine would perform on the 50°C/52°C. I too couldn't hear much more noise over the telephone line. Also note that the 8790 at 69°C here didn't throw on the CAPS and NUM lock lights as yours seem to do.

When he removed the cushion the temperature started to fall rapidly from 69°C, about ~9°C in say ~2-3 minutes to 60° and then further down...


Gary, I would say that your machine has some huge unnormal problems, since...

a) ...it shouldn't flash on the CAPS and NUM lock lights when on high heat...

b) ...it also should fast cool the machine down to normal parameters without making too much additional noise...

c) ...the cooling fans of the machine shouldn't run one hour after you shut down the system or something like that (e.g. what you reported above in this thread...).

Gary I think you should send the machine definitely back for an exchange, so you get another well behaving unit!!!


PS. I believe my Clevo buddy has now a burned cushion...

Hmm

All this talk about fan noise and such, does put teh thought into ones head of finding higher quality replacement fans to "mod" these heavy beasts. I noted in this thread talk about the fans being custom made for clevo, so I gather finding similar size, power req't, output fans is not a trivial matter. Since Clevo is in the business of mass producing these things to make money, its not too much of a stretch that they are using as inexpensive units as possible that meet a minimum spec. Anyone know of a replacement source for these things? I know I'd be willing to pay much more than a clevo to buy higher performing/more quiet fans if I knew of a reliable alternative...

I just finally made the plunge and ordered an 8790 equivalent. I went the route of a non-sager due to a couple of factors: 1) I wanted a 100% iron clad guarantee of no dead/stuck pixels, and I was willing to pay for it; 2) I just knew if I bought a Sager, the very first thing I would do is pull it apart and replace the thermal pad between the heatpipe and CPU with a premium thermal cmpd, and I'd fear voiding Sager's warrantee in doing so (hard for me to hold off tinkering with things);and 3) I've seen some of the paint jobs people have been getting on similar notebooks and thought I'd finally treat myself to a totally frivolous excess (I promise to post pics..). Time will tell whether or not the premium $$ I spent is worth it or not. FYI I ordered today, 3/26, initial expected ship date 4/25 (which according to my calendar is a Sunday, so little more than an automated "guessdate"). My current situation made the $$ and the extra delay less of a factor. The countdown begins...

- Card

I don't own a Sager but I do have a Gateway 600 that I had problems with it overheating and shutting down. Found out that the shipping people weren't very kind to the box during shippment and had dropped it hard enough to knock the heatsink far enough away from the processor to stop it from cooling correctly. Haven't had any problems since with it. I even overclock the 7500 to 290mhz and ram to 200 effective 400mhz. I am hoping to get my company to approve me getting a new notebook and would seriously like to get the 8790.

My 8790 locks up after about 30-45 min of playing Call of Duty and the Farcry demo.... my fans never do kick into turbo mode and the lights dont blink... I have been using modmeter and I cant tell what temp it locks up at as it only locks up when I play in full screen mode. By alt tabbing while gaming, im guessing its in the low 60's

I wish there was a way for me to set the fans to kick on earlier, so that I could play a game for longer than 45min... I dont notice noise when I am really into a game.

I got worse RAM!!!

Check out my spd.txt dump:

CPU-Z version 1.21
Memory Modules Serial Presence Detect (SPD)

Module #1

General
Memory typeDDR-SDRAM
Manufacturer (ID) (FFFFFFFFFFFFFFFF)
Size512 MBytes
Max bandwidthPC3200 (200 MHz)
Part number
Attributes
Number of banks2
Data width64 bits
CorrectionNone
Registeredno
Bufferedno

Timings table
Frequency (MHz)166200
CAS#2.53.0
RAS# to CAS# delay33
RAS# Precharge33
TRAS#78


Module #2

General
Memory typeDDR-SDRAM
Manufacturer (ID) (FFFFFFFFFFFFFFFF)
Size512 MBytes
Max bandwidthPC3200 (200 MHz)
Part number

Attributes
Number of banks2
Data width64 bits
CorrectionNone
Registeredno
Bufferedno

Timings table
Frequency (MHz)166200
CAS#2.53.0
RAS# to CAS# delay33
RAS# Precharge33
TRAS#78

I don't think that's true. I have a 8700@3.4ghz and my CPU-Z RAM report is the same as $tack$. It's a possibility that different RAM was used during different production periods. I don't know when mine was made but I recieved it on the 18th.

Retry your memory diagnostics with the CPU-Z 1.20a version, which might be possibly more stable here.

Also note, that not all SO-DIMM vendors do programm the SPD-EEPROM (Serial-Presence-Detect-EEPROM) of the SO-DIMMs in a well behaving manner. This is sadly said still the case for many No-name- and so called OEM-memory modules (which indicates a poor SPD-EEPROM handling by the memory modul manufactor).

In other words, in your case the memory modules are not worse, but the manufactor or source of the modules didn't wrote in the manufactor ID and part numbers into the SPD-EEPROM of the modules. - Or CPU-Z can't determine these value entries for your modules...

Less aggressive CAS timing = worse, i.e. lower threshold, RAM.

The other items are the 'icing on the cake' hallmarks of no-name memory.

sigh.

This would partially explain my less than anticipated benchmark scores (in other thread.)