Point of View 7300GT gfx card recap

[mariushm]

Re: Point of View 7300GT gfx card recap

reflowing and reballing is nVidia's department.

New consoles are AMD...

[cheapie]

Re: Point of View 7300GT gfx card recap

Okay, deltaT of 3 degrees is quite bad ass. At least not-so-good ass. You are insane man.

It's an AMD Athlon II X2 270, if you're wondering. It's got an Arctic Cooling Alpine 64 Plus on it, which, thanks to the FDB fan, I can run at full speed without it getting noisy. I think I have the motherboard's "Smart Fan" settings at a target temperature of 50 degrees (C), and a minimum fan speed of 50%. I have the cpufreq governor set to "conservative", so it only runs at 940 MHz when idle. It can be interesting to set the governor to "powersave" (locks it at 940 MHz) sometimes, just to see how well various programs run at that speed.

I can't stand ATI's quirky drivers and lousy TV-Out support.

I'll admit that AMD's drivers are no good, but I've never had any issues with the HDMI out. It works just as well as the others for me, and I've had two cards with it.

As for those drivers, once I got a Radeon HD 7790 and installed it. The 7790 has its own driver, so I installed that. The first thing that I noticed was that my mouse pointer was about four times its original size. Everything else worked fine, though. Later, I installed my 6570 as well (so I could use OpenCL on both at the same time). I did not connect them with CrossFire or whatever, as they're too different, and my 6570 doesn't support it anyway. When I turned the computer on, it crashed. I had to remove the 6570 (leaving the 7790 installed), remove the driver, reinstall the 6570 (so both were installed again), and install the generic drivers. Those drivers worked with both cards, so I'm not sure why the separate 7790 driver even exists.

[momaka]

Re: Point of View 7300GT gfx card recap

Just because newer graphics cards are more powerful doesn't mean they are planned to break. That is silly.

It's not silly at all. I don't care if AMD came up with a die that can withstand 9999999C... those BGA joints that hold the chip down to the PCB will still break when the delta T from the thermal cycling is too great. And the hotter the chip is allowed to run, the greater the delta T. Also, the lead-free solder is NOT helping it any either!

Most hardware nowadays is built like shit. No considerations at all for proper cooling or any other features to make the device last.

So yes, like mockingbird, I anticipate that the next Xbox will indeed need lots of reballing and reflowing. Maybe nowhere as bad as the older 360s, but still enough to keep repair businesses happy.

[quote=reflowing and reballing is nVidia's department.
New consoles are AMD...[/quote]
Doesn't really matter. What failed most of the time on the original Xbox 360 was the GPU BGA not the CPU. And the GPU was made by... ATI/AMD. Not really ATI/AMD's fault though. The GPU heat sink was just plain shitty.

[Th3_uN1Qu3]

Re: Point of View 7300GT gfx card recap

reflowing and reballing is nVidia's department.

Where have you pulled that from? Feast your eyes on this... http://www.servicepoint.ro/istoric.php

[dj_ricoh]

Re: Point of View 7300GT gfx card recap

Where have you pulled that from? Feast your eyes on this... http://www.servicepoint.ro/istoric.php

woow

[trodas]

Re: Point of View 7300GT gfx card recap

shovenose -

Just because newer graphics cards are more powerful doesn't mean they are planned to break. That is silly.

I believe that "silly" is not noticing the whole industry goes with the design for the dump (eg. planed obsolesence). THAT is, my fried, silly.

Even "silly" gal can figure that out - check this video out and edicate yourself, you gonna need that:
http://www.youtube.com/watch?v=9GorqroigqM


PS. Checking question. What do YOU think? Do industry use Teapo, Ost, G-Suxxon, Fuhjyyu and similar caps for what reasons?

[mariushm]

Re: Point of View 7300GT gfx card recap

Where have you pulled that from? Feast your eyes on this... http://www.servicepoint.ro/istoric.php

It's a bit biased... Romanian site, Romanian customers, with low wages, majority bound to buy budget laptops that aren't necessarily designed to last... or used laptops...

Fact is AMD/ati cards don't really have a history as much as nvidia has (all that fiasco with weak underfill and poor solder on the gpu chips)

[RJARRRPCGP]

Re: Point of View 7300GT gfx card recap

Fact is AMD/ati cards don't really have a history as much as nvidia has (all that fiasco with weak underfill and poor solder on the gpu chips)

Is that why a Radeon 9800 Pro I got second hand kept giving a corrupted BIOS screen and only work correctly intermittently.

[tazwegion]

Re: Point of View 7300GT gfx card recap

Is that why a Radeon 9800 Pro I got second hand kept giving a corrupted BIOS screen and only work correctly intermittently.

Hard to say... my R9600XT has been in use since new and now gets a harder time in the hands on my gamer son still without issues.

[momaka]

Re: Point of View 7300GT gfx card recap

Is that why a Radeon 9800 Pro I got second hand kept giving a corrupted BIOS screen and only work correctly intermittently.

No, that's just the RAM BGA going bad. Both Radeon 9500/9700 and 9800 had this problem. Not sure if this was a screw up on ATI's part making the RAM chips run too hot or a manufacturing fault. Those cards used leaded solder so they should not have had these issues. A reflow fixes them most of the time, though.

The GPU of the Radeon 9600 uses a newer die that runs much cooler, and also slower but cooler-running RAM so that's why it won't have any heat issues. However, funny thing is, the Radeon 9500 still outperforms the 9600 despite being older an a "lower" model.

[Wester547]

Re: Point of View 7300GT gfx card recap

No, that's just the RAM BGA going bad. Both Radeon 9500/9700 and 9800 had this problem. Not sure if this was a screw up on ATI's part making the RAM chips run too hot or a manufacturing fault. Those cards used leaded solder so they should not have had these issues. A reflow fixes them most of the time, though.

The GPU of the Radeon 9600 uses a newer die that runs much cooler, and also slower but cooler-running RAM so that's why it won't have any heat issues. However, funny thing is, the Radeon 9500 still outperforms the 9600 despite being older an a "lower" model.

That's because the Radeon 9500 (and Pro) was essentially a clocked down version of the R300 (Radeon 9700/Pro) with its memory interface bus width cut in half and the same for the amount of (locked) rendering pipelines on the 9500. The 9600 only had half the amount of rendering piplines (for texture mapping, vertex and pixel shaders, and rendering otherwise) and lower memory bandwidth. The 9600 Pro did have somewhat superior memory bandwidth over the 9500 Pro, though, but that only really helped it in synthetic 3D benchmarks (and even then it was largely outclassed by the 9500), which doesn't mean much of anything (especially considering how much nVidia cheated with their drivers back then, in order to make their GeForce FX cards look "comparable" to the 9500-9800 series). It could help a tad in games that use more traditional methods of rendering (texture operations) rather than shaders, but even then those were starting to be phased out.

If you read the old reviews of the 9600 Pro and compare it to the 9500 Pro, there really isn't a single situation (3D rendering in games) where the 9600 Pro does better, much less even equal the 9500 Pro. The R300s were all so far beyond their time that even within 4 years of their release, they could still play the newest games at decent settings. I didn't realize the 9600s didn't run as hot, though - that's definitely a good thing (in my experience, however, the R300's BGA RAM chips only run very hot in 3D rendering and aren't even warm to touch in 2D operation). I'd guess the R300 fails because its core and RAM chips are of the BGA package and it isn't -that- tolerant to large temperature swings (flexing, though still much moreso than leaded solder). Some of the stock models also had shoddy thermal paste and an ADDA hyprobearing fan that would eventually seize (expediting the failure).

I would say nVidia's problem with the underfill is a little more perturbing than even the worst of ATi/AMD's driver, issues, catalyst control crap, and whatever else. nVidia knew about the underfill issue with the GeForce 6xxx-9xxx series of cards and lied about it through their teeth. It's too bad that problem plagued upon the 8800 GTX/GTS cards as well because while those gigantic beasts ran very hot they, like the R300s, were beyond their time too as far as performance goes.