With new cards coming out...

Will the resellers help those of us who have put up with 880s get new cards? I mean it's obviously something we are going to have to pay for which is already not fair to us since it was nVidia who messed up but I want a warranty on the cards when I'm dropping that kind of cash.

All of us who have the 880M will undoubtedly be looking for a solution to get new cards as everyone knows that these are just plain broken and the "solutions" to the issues are far from acceptable. Let's face it, nVidia screwed all of us here who bought these expensive top tier machines by giving us an inferior, unstable, and very hot video card and my 5k would have been better spent on the older 780M. That's just really bad...

Sent from my HTC One_M8 using Tapatalk

 

If you can prove your cards are broken under warranty you might get 980Ms as they'd be the new flagships... but I don't think saying "hey reseller I have 880Ms, give me new cards cuz these are broken" work.

Though in your case since your cards still do not perform well at their designed clocks you have a solid reason; though Sager is less accommodating than say.. alienware.

 

Alienware wont be going around swapping cards.

 

Obviously I'm willing to pay for the upgrade, I just want to be able to purchase it from Sager or a reseller to have a warranty. Dell would do that, even if they gouge on the price.

880Ms are a special case since the cards are all defective. I would have already sent my machine back again if it didn't cost so much to ship but I'm not going to keep paying 240 for shipping when I'm just going to get the same broken cards anyway.

You can guarantee that people are going to be truly throwing a fit over these busted cards if the 980M comes out with no problems. That will mean that both the card before and after were great while this generation is an overheating disaster.

We shouldn't all get screwed and told "too bad" especially if we are willing to pay for the upgrade.

Sent from my HTC One_M8 using Tapatalk

 

I can't speak for others but we have our upgrade policy that allows this sort of thing to be done and Sager is more than willing to accommodate, providing that the cards are compatible. That information isn't yet available since the cards aren't released and being used. Once that is determined and the compatibility is there, you won't catch us saying no you're out of luck.

 

Note that the EM and SM machines never got the 7 or 8 series so nothing is guaranteed.

 

Well one of the test machines was a P370SM-A so there's still hope.

Thanks for giving me some hope Derek, I really hope that upgrading becomes an option on the table for my machine.

Sent from my HTC One_M8 using Tapatalk

 

I have an 880m and I've not noticed any serious problems.
Why are all the cards defective?

 

Too much to go into here. Go look at the Pidge from nvidia thread in the gaming hardware section or the various 880M threads in the alienware section. Stock core throttles down to as low as 888MHz when the card isn't even running hot (i've seen it happen at 79C). 780M can run 880M clocks with no throttle and lower heat even though it takes more voltage for those cards to reach and hold our clocks and 880M is just a highly binned 780M with 8GB of RAM which means they should be superior but they are not. Definitely broken. And that's just the tip of the iceberg. I had to have mine replaced once already because the slave flipped out one day, crashed and disappeared then randomly came back. Good thing xotic took care of me.

Sent from my HTC One_M8 using Tapatalk

 

No luck here... I built my machine from a barebones I bought. Highly doubt there would be warranty on that.

 

I've not encountered any of the mentioned problems with my 8800 but then I've not tried to overclock.
I had some minor stability problems with some of the older drivers though.

 

You have the problems, you just aren't monitoring for them. These aren't variables, they are known facts. Stock vbios is broken and sets the card to throttle too much and the modded vbios that allows it to boost all the time releases the ridiculous heat output of the card. Don't live in denial, install afterburner and watch the OSD or use GPU-Z to make a log and you Generic Log viewer to make a chart showing the dips and spikes in the core. Trust me, I didn't think I had the problem either until I went and read here and checked for myself.

Sent from my HTC One_M8 using Tapatalk

 

Have to say, with my 880M I'm not really seeing much issues on my stock settings.

I've seen some occasional drops in the GPU to 836Mhz, and once for a second down to 770Mhz, but that timed directly to gameplay when I was browsing my inventory where it wouldn't have utilized the GPU anyway.

For the duration I was running around the country side and killing monsters, it stayed pretty solid at 993Mhz core with 1250Mhz clock.

GPU temps never exceeded 70c.

GPU utilization peaked at 97%, averaged around 85%.

Do I see drops? Yup, but they seemed consistent to the gameplay for me at least.

Game I tested it on was Skyrim, 1920x1080 ultra everything.

I should note that I have a single 880M in my 8258, and everything I've read with the 880M issue seems to be tied either to overlocking them or SLI mode.

 

You're right, but I've got a few graphical mods on Skyrim, and some places, fps dips dramatically, so I can't stand the stock vBIOS. On the other hand, svl7's vBIOS makes the card run way to hot, so I don't know if I can find a balance between the two.

 

Hum. Just a thought, but since you have a lot of custom mods, could it be that the dips in your GPU are concurrent with heavy I/O access while it's loading in new areas?

If the I/O is bottle necking, it's possible it coincides with a GPU drop since at that point the GPU obviously won't be utilized.

Oh, I also have Skyrim HD 2K textures, Static Mesh Improvement, HD Fire effects, W.A.T.E.R, SkyUI, Skyfalls + Skymills, Ruins Clutter Improved, Magic Runes HD, Color Map Markers HD, Burnt Corpses, Better Dynamic Snow, and enhanced blood textures along with dozens upon dozens of gameplay alteration and content mods.

So I'm running Skyrim a fair bit modded as well.

I should also mention I'm running Win 7 Pro 64bit with the GeForce 340.52 driver, and the 7200 1TB RPM mechanical drive.

Of note, there's been a lot of unconfirmed reports of graphic stuttering and GPU drops with Windows 8, not just the 800 series cards,but apparently affecting ATI as well as older generation GeForce cards.

You can find info on searches with windows 8 graphic stuttering

 

I've got all those mods as well as some lighting ones. The FPS drops might just be because of the fact that some areas are more crowded than others. When I play on svl7's vBIOS, my temps stay steady at 93 degrees but no throttling seems to be apparent. I'll look into the stuttering issue, since I am running Windows 8. As for drivers, I'm on the 340.76, and for some reason, seems to give me slightly better temperatures.

 

Are you absolutely sure it's not hitting 94C and throttling in the 400s every now and then?

Sent from my HTC One_M8 using Tapatalk

 

I've prioritized temps at 93 degrees, and it seems to stay even at 93, and occasionally dips to 92. If I set a -80 on the core and 400mhz on the memory, I get stable temps at 92/91 degrees. I believe I am just on the edge of throttling, but haven't seen temps of 94 as of yet.

 

Skyrim has been known to drop FPS a lot for no reason. More than once I'll be getting 35-40 fps in one area using something like 40% load on both cards... depends on the area and the mod usually, but it isn't actually your cards' faults. The rest of the game usually runs at 120-150fps if I don't adaptive vsync it half refresh rate for me on stock 780Ms.

Unless you're on a single card and you're actually running out of power, in which case that might be the reason XD. If that's the case take what I said above with a grain of salt. If you wanna see your GPU util, download and run Playclaw 5 & set up the in-game overlays. I always check my GPU util and filled memory buffer and temps in real-time in-game. Same for my CPU

 

Lol you are far braver than I. Those temps are too much for me to handle. I dont like above 85.

Sent from my HTC One_M8 using Tapatalk

 

Haha, I would like better temps but unfortunately, the GPU always maxes out at 93, unless I'm playing something that only uses around 80%, then it stays about 10 degrees cooler.

 

Its ridiculous. My cards are acting up again too. So over this. It's nice to know that this is the quality you get when you drop serious dough..........

Sent from my HTC One_M8 using Tapatalk

 

Ethrem,

We are currently working with Sager on a resolution, will update you when we hear back and sorry for your troubles it is not typical!

 

I would like to know as well!

 

Please contact me through email to discuss resolution

 

You guys are awesome. It was another really bad day realizing that my cards are acting up again. I want to make it clear that I'm not upset with anyone here except nVidia. You guys and Sager both seem to be doing what you can to accommodate those of us that have issues with these cards and every time I've contacted you guys about anything with my machine, I received a response back within the same day and a professional response every time as well.

It just makes me mad that nVidia seems to really not have any care at all about the situation. In the entire forum there is one set of cards that do not seem to throttle at all OR get excessively hot. Having another pair starting to go out is just nuts. First set I can chalk up to bad luck but two? That is poor engineering somewhere down the line.

Really nVidia just needs to recall these cards and find a suitable replacement for them. But that won't ever happen because it would hurt their bottom line so they leave the mess to the OEMs and resellers to sort out. It's not right!

Sent from my HTC One_M8 using Tapatalk

 

As I also have an 880M, I'd like an option as well.

In fact, I'm sure anyone with a GeForce 800 series card would like this.

If Nvidia would give anyone who purchased a laptop with a GeForce 800 series a fairly nice discount on a 900 series nvidia card for the laptop, it'd be a nice thing.

I know my 880M isn't a problem and I'm happy with it, but if I had a chance to get a 980M at a fairly reduced price, I'd probably pick one up for when the 880M does go flakey.

Thoughts?

 

My opinion is that if nVidia were to allow some sort of recall/exchange program, the 980M would likely be excluded from that scenario and it would be 970M cards since they outperform the 880M by a good margin.

I think that would be a fair resolution but nVidia doing that is extremely unlikely.

The good news is that they've been testing the new cards in existing machines so Clevo could potentially use the same models instead if launching a new line which would mean that Clevo and Sager would have the option of replacing the problematic cards on a case by case basis without having to replace the entire machine.

I'm still extremely irritated that I can't game yet again though and it makes me extremely nervous trusting UPS to cart an expensive machine safely across the country multiple times too. The last time I was worried about it the whole time from shipment til I got it back.

The 870M model doesn't suffer from any of the 880M issues either from what I've seen. It's almost like nVidia did this on purpose to make sure our 880M cards couldn't stand up to Maxwell but they went too far and made sure they couldn't even stand up to the older 780M.

Sent from my HTC One_M8 using Tapatalk

 

I've purchased an 880m laptop from xoticpc about 2 months ago I'm curious how this will play out. I've been mum on this subject as I've learned more as it developed. I was hoping for a fix from Nvidia but if that is not in the pipeline, I'm very curious as I said on what is going to happen. This is my first time buying a laptop such as this. The cards all seem to be borked at stock settings.

 

Well over the course of these few months, I'd say the new drivers have made the 880m slightly more bearable. At least for me anyway.

 

Agreed.

When I first had my 880M, while the performance seemed to be 'on spec' the overall thermals were far too high.

The progression through the drivers seem to have kept the performance and greatly help the thermals.

I hear that people who have SLI 880M's or have attempted to overclock them still run into severe roadblocks. I feel for them, but not OC mine nor having an SLI system, I'm spared those issues.

The latest drivers released have improved stability even further as well, and may help the above issues, will be curious on testing with it that people provide.

It seems pretty steady on performance,the thermals stay relatively low for the power, and overall for the 880M, barring any really wonky behavior yet to be found, I'm very happy with it.

 

To be honest, I'm actually not worried about thermals, albeit how ridiculously hot these chips run. E.g. I just ran a game for around 3 hours today, no throttling whatsoever with the GPU steady on 93 degrees. Like you, I am satisfied with the performance though. These 'issues', I seem to be able to live with now.

 

I was also happy with the drivers making things progressively better but I have an actual hardware failure and this is the second time.

First time, my slave card up and disappeared from the system. GPU-Z showed it's clocks but it said Revision FF and had no temperature or other monitors when I started up HWiNFO64, didn't show up in nVidia control panel, etc. Then it randomly started working again but it was crashing games. Sent the machine off to Sager to replace the slave, they opted to replace both.

Now these new ones were working great. 87C max, master averaging 79C and slave averaging 82C then I started getting artifacts the other day and it started crashing. Now no matter what I do, the cards will crash eventually. It could be 10 minutes or it could be 4 hours. My guess is that it's a vram issue since it can take so long to trigger. I also had a few incidents where the cards locked at 993MHz core yet the frame rate dropped from the 90s to the 20s.

This happened with three different drivers and I was cleaning them properly with DDU each time.

I said awhile ago that I was willing to live with the cards and was just disappointed with the crippled performance but I obviously can't live with broken cards so that's what sparked my emailing xotic and asking if we have any other options. I can't very well sell a broken machine in good conscience and being out 5k is about as unacceptable as having to have my cards replaced every month (its only been a month since I got my RMA done) and having to worry about shipping my machine back and trusting ups every time not to lose it or damage it and cost me even more money. I know all too well when things get broken in shipping it becomes a sticky mess figuring out who was at fault and I've been screwed by UPS before.

If your cards are working, I wouldn't worry about it now. It's pretty obvious that Xotic and Sager will take care of those of us who have issues.

Sent from my HTC One_M8 using Tapatalk

 

your experience has really kind of steered my away from buying a Sager in the future.

 

Why? Sager hasn't been a problem at all so far. Dell would have sent out at least one bad technician by now who would have seen it as an intermittent issue and not replaced the cards. Sager still replaced the cards last time. With Dell you have to have a catastrophic failure before they will do anything about it.

My experience with their warranty process has been a good one so your statement confuses me.

They are essentially fixing what nVidia refuses to.

I have had horrible experiences with Dell so it's actually refreshing to deal with a company that actually cares about their customers. If I sent an email to Dell about the issue, it would be many hours before I got a response and it would be something like "Sir, please to making sure your computer is using Dell drivers" even though I would have already told them that I tried all the different drivers in my first email and it would take multiple exchanges to clarify my issue. If I called them on the phone, it would be over an hour of working through the language barrier before my ticket would be submitted and when the technician would be sent to the house, it would be luck of the draw if they do anything about it at all or say they don't see any issue and have me send it to the depot for extended testing and send it back to me with parts not connected inside the machine, leaving me to fix them myself.

My machine already went to Sager once and they didn't screw anything new up, just replaced the cards, and put the computer back to default settings including putting the SSDs back in RAID and putting a fresh Windows 8.1 install on there. Noticing that I ordered my machine with RAID0 and setting that back up is what I'd call thorough, its not something I would expect from the repair center to pick up a small detail like that.

Sent from my HTC One_M8 using Tapatalk

 

Not so much the company and their policies but the quality of the machine. I have not had any of these issues with my cards so far which is worrying from a quality standpoint.I understand that you may feel the blame lies on nVidia, but apart from the known issues inherent with the 880m these cards have been running pretty much flawlessly.

 

There are plenty of cases of defective cards in the Alienware forum too though. It's the 880M, not one particular manufacturer. These are reference boards made how nVidia designed the specifications. The only real difference between these and the Dell cards are the memory used and the vbios.

Generally speaking, Clevo cards are held as the best cards because they actually follow the MXM spec and can be used in any MXM compatible machine while everyone else makes spec changes that can break compatibility.

Sent from my HTC One_M8 using Tapatalk

 

I'm with you on that, but it just feels suspect to me.

 

Anything is possible but keep in mind that the only truly working 880M cards on the forum are Clevo cards. sa7ina has a set that don't drop below 954MHz after being stressed with Valley and the temps are a lot lower too. Even being in a P570WM, there shouldn't be a 10C temperature difference at load when we are both using the stock IC Diamond.

Sent from my HTC One_M8 using Tapatalk

 

Not to mention the fact that the cooling for the P370SM I have and Ethrem's P377SM-A (or is it P370SM-A he has? Same difference really) is basically the same, and under most circumstances I don't crack 72 degrees at stock or 80 degrees OC'd to 950/6000 on my 780Ms... and my room is *WAY* hotter and more humid than his. And he has gone through multiple 880M pairs too. If the cards are essentially the same and the cooling is essentially the same, the fault lies in the cards.

 

I was hoping to get a new heat sink for my machine. I looked around on eBay and discovered that my heatsink was for the 860m.hopefully a new heatsink might drop temps a bit more. I have already modded my back cover and printed less restrictive vents with the 3D printer, and that has eliminated the issue of throttling. Now all I need to solve is the temperatures.

 

Just make sure you get the thermal pads right as it makes such a difference.

 

You have an 860M heatsink? That seems odd. Do you have a picture of it? I would think Clevo would use the same heatsink for everything. The slave heatsink in my P377SM-A is marked P370SM I believe.

Sent from my HTC One_M8 using Tapatalk

 

I can explain what it looks like. It's like the 680m heatsink with the two pieces, except the memory piece has the top chopped off, leaving a huge void. Looks like Deadsmiley's heatsink. I can post a pic up later.

 

I just got a few pictures of it.


I cover up the void with some foil tape so it's no big deal. There used to be a plastic strip to cover it but I removed it.

 

Yeah I think mine is the same way. The heatsink fins are a mess on both. One of them has a stack of foam pieces to close the gap in the left side between the fins and the chassis lmao

Sent from my HTC One_M8 using Tapatalk

 

All the gaps on mine have been sealed up with tape one way or another. I also have a small strip between the fan and the memory heatsink, because there's a small gap there. Reduced my temps a little, so I'm happy. I haven't reached 93 degrees in 2 days.

 

My machine is going in for another RMA. Sager said they could downgrade to 780Ms but that I'd have to pay to do so because the 780M costs them more than the 880M does so I'm just going to take a third pair of 880M and hope that the third time is the charm.

I sent Xotic screenshots that I took from when I got the cards to now... It was a month to the day before the degradation happened on both cards. Its a good thing that I've gotten in the habit of monitoring my GPU performance or it would be much harder explaining that its not all in my head.

When I got the cards, I had a 20 minute Bioshock Infinite run with the average core being 980 between both cards and a linear rise in temps.

The chart from a month later, the master was averaging 963 and the slave was averaging 853 and the slave was 4C hotter than the master despite being 100MHz lower on the core.

That's some pretty serious degradation...

 

Can I rent two of your 880M returns? Not Ethrem's though.

 

Oh come now, you know you want to be unable to game for more than 20 minutes! Lol.

Sent from my HTC One_M8 using Tapatalk