since i have a bad eMMC chip can I use my warranty to return it for a new one?
[Q] Can I return my phone since it has a bad chip?
- Thread starter vtrinh93
- Start date
No because apparently HTC does not recognize that as an "official" problem. I already tried and they would not return it on that basis.
You could call tmo and say that its randomly rebooting or something..
Sent from my HTC Glacier using xda premium
Sent from my HTC Glacier using xda premium
That's def a fact. The sad thing is that if this were a computer with a faulty anything it wouldn't be hard for a customer base to push for a recall successfully. But with cell phone manufacturers they can successfully ignore us, the consumers.
Sent from my LG-P999 using XDA App
Sent from my LG-P999 using XDA App
You can if it bricks. Until then, I wouldn't worry about. I've been flash happy since march with no issues. I have the "bad" chip as well.
Sent from my HTC Glacier using XDA App
Yes, it's definitely a fact that this is not a problem at all, and has been blown completely out of proportions, mostly (as it usually happens) by those that don't understand anything in what they're passing a judgement on.
And if you happen to mention PCs - remind me, please, how many HDDs are failing when used? I'll tell you, if you don't know - that would be A LOT. It happened to me and my friend on brand new, 1/2 year old laptops. Both had their HDDs exchanged under warranty, and that's it. The same happens with the phones. What recall are you talking about exactly?
With PC's, just for instance/argument sake, the nvidia integrated issue between '08-'10 that affected dozens of production model HP's, Dells, Acers, and even some MacBook Pro variants. Severe overheating caused by a chip misdesign and horrible nonlead solder, causing massive recalls and a successful class action suit against nvidia, that was JUST settled this year. Not to mention the screen flickering issue on Asus G72/G73 laptops, a 2,000 dollar laptop.
On cell phones, lets consider for instance the LG G2x that has/had a DEFINITE hardware flaw in earlier productions that caused random reboots or noticeable screen bleed that was denied by LG. Or the MT4G screens that are horribly washed out when compared to other MT4G models produced at the same time. The eMMC issue is a VERY real issue, even more noticeable when they are taken apart and compared. The solder is jacked and the eMMC design is different, indicating it was a KNOWN issue that was covered up.
The way HTC looked at it, the phones are locked down and we aren't "allowed" to flash it within our warranty (which is asinine when it comes to hardware issues that weren't caused by rooting/flashing) and by the time any OTA updates caused it, they would be out of warranty. HDD's to a point are expected to fail and regardless were repaired under warranty with no issues, a luxury we don't have with cell phones. Not to mention these issues are WAY beyond a mere HDD that any person with a Phillips head could replace.
Sent from my LG-P999 using XDA App
On cell phones, lets consider for instance the LG G2x that has/had a DEFINITE hardware flaw in earlier productions that caused random reboots or noticeable screen bleed that was denied by LG. Or the MT4G screens that are horribly washed out when compared to other MT4G models produced at the same time. The eMMC issue is a VERY real issue, even more noticeable when they are taken apart and compared. The solder is jacked and the eMMC design is different, indicating it was a KNOWN issue that was covered up.
The way HTC looked at it, the phones are locked down and we aren't "allowed" to flash it within our warranty (which is asinine when it comes to hardware issues that weren't caused by rooting/flashing) and by the time any OTA updates caused it, they would be out of warranty. HDD's to a point are expected to fail and regardless were repaired under warranty with no issues, a luxury we don't have with cell phones. Not to mention these issues are WAY beyond a mere HDD that any person with a Phillips head could replace.
Sent from my LG-P999 using XDA App
Screen flickering and graphics chip overheating by design are design flaws. HDD failure is a component flaw. HUGE difference. And for argument's sake, eMMC is an equivalent for HDD, and not for anything else.
Moreover, as I already said, the "issue" has been blown out of proportion. Out of millions of MT4G sold, several hundreds (and I'm probably exaggerating here) went dead because of eMMC failure, and looking at the forum poll results, it's safe to assume that at least 1/3 of them had the "good" eMMC failed - which completely ruins your theory. Most, if not all, replaced under warranty. That percentage of failure in devices would most probably fall within QA demands for any reasonable electrical device quality standard. Again, what "flaw" are you talking about? This "bad" chip is sold in quantity of tens of millions and used across tenths of phones and other devices. This "issue" is real, as I said, only for those that don't understand a thing.
---------- Post added at 09:07 PM ---------- Previous post was at 09:00 PM ----------
And you can put attention that I'm not arguing other cases, which might be true and might be not. But this virtual "eMMC issue" got way too annoying to see mentioned all around the forums. There was a period when everyone would regard the "bad" eMMC as a "ticking bomb", suggesting not to root if it's found on the phone, completely forgetting to get the chance of failure in proportions. I'm glad this period is mostly over.
Moreover, as I already said, the "issue" has been blown out of proportion. Out of millions of MT4G sold, several hundreds (and I'm probably exaggerating here) went dead because of eMMC failure, and looking at the forum poll results, it's safe to assume that at least 1/3 of them had the "good" eMMC failed - which completely ruins your theory. Most, if not all, replaced under warranty. That percentage of failure in devices would most probably fall within QA demands for any reasonable electrical device quality standard. Again, what "flaw" are you talking about? This "bad" chip is sold in quantity of tens of millions and used across tenths of phones and other devices. This "issue" is real, as I said, only for those that don't understand a thing.
---------- Post added at 09:07 PM ---------- Previous post was at 09:00 PM ----------
And you can put attention that I'm not arguing other cases, which might be true and might be not. But this virtual "eMMC issue" got way too annoying to see mentioned all around the forums. There was a period when everyone would regard the "bad" eMMC as a "ticking bomb", suggesting not to root if it's found on the phone, completely forgetting to get the chance of failure in proportions. I'm glad this period is mostly over.
I'm asking because my keyboard freezes after the gingerbread update
Sent from my HTC Glacier using xda premium
Sent from my HTC Glacier using xda premium
"Freezes" = stops responding completely?
In that case, you can most probably return it, not because of the bad chip, but because of the freezing keyboard.
In that case, you can most probably return it, not because of the bad chip, but because of the freezing keyboard.
**not meant to be taken as argumentative**
I agree that HDD failure falls under the component column so to speak. HOWEVER an integrated *on board* eMMC is not a user replaceable component nor a component added after the board production like a HDD. An eMMC, furthermore, is not an equivalent for a HDD but rather an equivalent of an integrated BIOS chip (read as eMMC, most tech sheets for said motherboards take that liberty) which , if it fails, is a DESIGN flaw not a component level failure. Please forgo assuming that no one but yourself in these forums has a grasp on technology or its implementation on a "socket" level or initial design process.
As far as the other "tens of millions" of bad eMMC devices vs the "several hundred" you say have failed in comparison, your theory definitely would hold up if it weren't Extremely broad aand generalized. The bad eMMC units general quantity more than likely would come down to a production batch. Lest we forget also, those "tens of millions" of USERS with that batch of MT4Gs are not represented in ANY poll numbers from this forum or others as we utilize our devices far beyond the general "power users". We are a minority when it cones to this issue because we ACTUALLY flash/update our phones to a point where a design flaw would show it self. This does not mean a company should justify putting a product that is BOUND to fail for active users by saying we are to blame.
Sent from my LG-P999 using XDA App
I agree that HDD failure falls under the component column so to speak. HOWEVER an integrated *on board* eMMC is not a user replaceable component nor a component added after the board production like a HDD. An eMMC, furthermore, is not an equivalent for a HDD but rather an equivalent of an integrated BIOS chip (read as eMMC, most tech sheets for said motherboards take that liberty) which , if it fails, is a DESIGN flaw not a component level failure. Please forgo assuming that no one but yourself in these forums has a grasp on technology or its implementation on a "socket" level or initial design process.
As far as the other "tens of millions" of bad eMMC devices vs the "several hundred" you say have failed in comparison, your theory definitely would hold up if it weren't Extremely broad aand generalized. The bad eMMC units general quantity more than likely would come down to a production batch. Lest we forget also, those "tens of millions" of USERS with that batch of MT4Gs are not represented in ANY poll numbers from this forum or others as we utilize our devices far beyond the general "power users". We are a minority when it cones to this issue because we ACTUALLY flash/update our phones to a point where a design flaw would show it self. This does not mean a company should justify putting a product that is BOUND to fail for active users by saying we are to blame.
Sent from my LG-P999 using XDA App
1) The "general population" got the 2.3.4 update while well within warranty period, so if "excessive flashing" is the argument used for justifying the "bad quality" of the specific eMMC in question - it doesn't hold. All MT4Gs were flashed all across /system partition.
2) The company clearly states that rooting voids the warranty, and we do it knowingly. They don't test the phones or their components for excessive flashing cycles, and rightfully so. We choose to do what we choose to do, knowing that we're the only ones to blame if we screw up something. I believe that the time cleared the "rooted vs non-rooted" failures enough for everyone to understand that some eMMCs of both kinds are bound to fail, and if they are - they'll inevitably fail (as all eMMCs do, but in different periods of time), rooting or no rooting.
3) You should probably remember that eMMCs were new by that time, and I agree that it's most likely a faulty batch, possibly one of the first batches. So? If devices will be replaced upon failure - even failure caused by rooting (which HTC did) - I don't see any case for claiming "flaw by design".
4) eMMC is not an equivalent of BIOS, simply because it's a storage device. Being soldered to the board doesn't make it "design flaw". Design = creation of board and making sure all the components are used within their specific requirements. If a component fails - soldered to the board or not, it's a component failure. Let me give you another analogy: let's say if Atom CPUs are failing when overclocked, is that a design flaw of whoever made the netbook with them? They're also soldered to the board. How does a complaint like "These guys are bad, they're selling us a flawed design that is sensitive to overclocking, even though it's against the warranty to overclock, we should claim a recall" sound? Doesn't sound fishy to you? I bet it does. But why do you think that "These guys are bad, they sold us a flawed design that is sensitive to rooting, even though it's against the warranty to root, we should claim a recall" sounds any better?
5) You're proving my point - if any, the "failed eMMC" users are highly over-represented in the forum poll. I can argue that almost any power user that knows how to root, and had his phone fail after rooting, would come to XDA to seek an advice - and that's probably going to be much more closer to the truth.
6) You prove again that you don't understand the issue itself, right after asking me to assume that it's otherwise. The eMMC is not "bound to fail" when rooted. Specific percentage of the eMMC chips, both "good" and "bad", is BOUND TO FAIL EVENTUALLY WITHIN A SHORT PERIOD OF TIME - rooted or not rooted. And it will ALWAYS be like that - a small percentage of any produced eMMC WILL FAIL shortly after being used. These are essentially the same chips used on SD cards, and the rate of failure is most likely similar, and as you probably know, SD cards fail sometimes. No amount of quality control can prevent statistical spread of silicon lifespan.
7) If you really want to argue "rooting sensitivity" as fault, then you get the strongest possible argument against you, which will probably explain the reason why I dislike this thread:
The manufacturer, as I explained earlier, should design the phone to have all the components working within their specs. To root the phone, you intentionally severely violate the spec of one of the eMMC - power-cycling it without preparation to begin with, and creating a race condition to prevent it from being write-locked when it's powered back on. This violation somehow escapes your "faulty design" argument, but it shouldn't - the faulty design in this case is created by the users that are rooting. And this whole idea of "I'm going to root, even though it voids warranty and possibly stresses one of the system components beyond any possible tests, violating its specifications, and if it sometimes fails - it means I'm screwed by the manufacturer and should get compensation" is well along the lines of "I'm going to buy a hot coffee, and if I spill it - I'll blame and sue McDonalds" way of thinking, which I despise and fight wherever I find - it's not that I like manufacturers.
Anyway, too much time wasted on nothing, not going to argue anymore. If I didn't convince you, and I probably didn't - you can think whatever you'd like to.
2) The company clearly states that rooting voids the warranty, and we do it knowingly. They don't test the phones or their components for excessive flashing cycles, and rightfully so. We choose to do what we choose to do, knowing that we're the only ones to blame if we screw up something. I believe that the time cleared the "rooted vs non-rooted" failures enough for everyone to understand that some eMMCs of both kinds are bound to fail, and if they are - they'll inevitably fail (as all eMMCs do, but in different periods of time), rooting or no rooting.
3) You should probably remember that eMMCs were new by that time, and I agree that it's most likely a faulty batch, possibly one of the first batches. So? If devices will be replaced upon failure - even failure caused by rooting (which HTC did) - I don't see any case for claiming "flaw by design".
4) eMMC is not an equivalent of BIOS, simply because it's a storage device. Being soldered to the board doesn't make it "design flaw". Design = creation of board and making sure all the components are used within their specific requirements. If a component fails - soldered to the board or not, it's a component failure. Let me give you another analogy: let's say if Atom CPUs are failing when overclocked, is that a design flaw of whoever made the netbook with them? They're also soldered to the board. How does a complaint like "These guys are bad, they're selling us a flawed design that is sensitive to overclocking, even though it's against the warranty to overclock, we should claim a recall" sound? Doesn't sound fishy to you? I bet it does. But why do you think that "These guys are bad, they sold us a flawed design that is sensitive to rooting, even though it's against the warranty to root, we should claim a recall" sounds any better?
5) You're proving my point - if any, the "failed eMMC" users are highly over-represented in the forum poll. I can argue that almost any power user that knows how to root, and had his phone fail after rooting, would come to XDA to seek an advice - and that's probably going to be much more closer to the truth.
6) You prove again that you don't understand the issue itself, right after asking me to assume that it's otherwise. The eMMC is not "bound to fail" when rooted. Specific percentage of the eMMC chips, both "good" and "bad", is BOUND TO FAIL EVENTUALLY WITHIN A SHORT PERIOD OF TIME - rooted or not rooted. And it will ALWAYS be like that - a small percentage of any produced eMMC WILL FAIL shortly after being used. These are essentially the same chips used on SD cards, and the rate of failure is most likely similar, and as you probably know, SD cards fail sometimes. No amount of quality control can prevent statistical spread of silicon lifespan.
7) If you really want to argue "rooting sensitivity" as fault, then you get the strongest possible argument against you, which will probably explain the reason why I dislike this thread:
The manufacturer, as I explained earlier, should design the phone to have all the components working within their specs. To root the phone, you intentionally severely violate the spec of one of the eMMC - power-cycling it without preparation to begin with, and creating a race condition to prevent it from being write-locked when it's powered back on. This violation somehow escapes your "faulty design" argument, but it shouldn't - the faulty design in this case is created by the users that are rooting. And this whole idea of "I'm going to root, even though it voids warranty and possibly stresses one of the system components beyond any possible tests, violating its specifications, and if it sometimes fails - it means I'm screwed by the manufacturer and should get compensation" is well along the lines of "I'm going to buy a hot coffee, and if I spill it - I'll blame and sue McDonalds" way of thinking, which I despise and fight wherever I find - it's not that I like manufacturers.
Anyway, too much time wasted on nothing, not going to argue anymore. If I didn't convince you, and I probably didn't - you can think whatever you'd like to.
Last edited:
With all that's been said i hope you don't just go about in returningbthe phone expecting to get a new phone with good chip.. all this bull **** is about revenues. They are designed to make manufacturer money. Why make all the chips good, if they do then no body would want to buy another phone. Anyways, the odds of you receiving another phone with a good chip is slim to none. My girlfriend had to return her phone with stock Rom because of the buttons failure. Before she returned it I checked and she had the.bad chip. Next day when she gor the new phone from T-Mobile I checked again and it still had the bad chip. All am saying is that the expected life of good or bad chip is clearly unknown. Your bad chip might last longer than someone with a good chip. I have the bad chip as well and I've spoken to HTC and they all act like they've never heard of a bad emmc chip. I've flashed ics Rom, sense roms. And still going strong. But am currently thinking about going back to stock Rom and just stick with it.
Sent from my HTC Glacier using xda premium
Sent from my HTC Glacier using xda premium
Both of you (Jack & Jeremy) good argument, BUT its misplaced.
Jeremy the whole issue about the eMMc is subjective which can NOT be replicated under any controlled conditions even now. But you are correct to change the components in the SoC for general publicis not practical. Even for the manufacturer the cost of part and labor would outweigh the cost of band new motherboard which is easy to replace. So should the company be held accountable? No, technically HTC or any other manufacture would never admit an issue exist which would impact its integrity, quality of its product as whole.
Jack don't trust the poll as its manipulated and no one in our Glacier forum suffered failed partition update with SEM04G chip as mostly its reported by M4G2DE chip users. Once again don't look at the poll as I said its manipulated. But yes that does NOT mean SEM04G chips won't die also return same fail-pu flag, so far I haven't seen a valid case of it here.
Keep in mind both of you that no chips are created equal, even today. Because during production each batch has its own defect rate which depends on the quality of the chip's wafer. But the biggest killer of any media storage device is time. Which is heavily dictated by user's daily activities as it comes with tear and wear. Which would directly impact the chips projected lifespan.
You also have to understand how flash media works specially on our own device. Glacier/MT4G is formatted using SLC-EUDA. Which is where the problem lies as you can't actually do low level format to the chip. Which would normally fix the corrupted partition table as the quota for the media would be defined normally by doing BINFS format and re-declaring the cylinders you can fix it. But we can't so we only have access to FS format on already pre-defined partition which is already declared as it can't be done again other than replacing the particular eMMc.
Sent from my HTC Glacier
Jeremy the whole issue about the eMMc is subjective which can NOT be replicated under any controlled conditions even now. But you are correct to change the components in the SoC for general publicis not practical. Even for the manufacturer the cost of part and labor would outweigh the cost of band new motherboard which is easy to replace. So should the company be held accountable? No, technically HTC or any other manufacture would never admit an issue exist which would impact its integrity, quality of its product as whole.
Jack don't trust the poll as its manipulated and no one in our Glacier forum suffered failed partition update with SEM04G chip as mostly its reported by M4G2DE chip users. Once again don't look at the poll as I said its manipulated. But yes that does NOT mean SEM04G chips won't die also return same fail-pu flag, so far I haven't seen a valid case of it here.
Keep in mind both of you that no chips are created equal, even today. Because during production each batch has its own defect rate which depends on the quality of the chip's wafer. But the biggest killer of any media storage device is time. Which is heavily dictated by user's daily activities as it comes with tear and wear. Which would directly impact the chips projected lifespan.
You also have to understand how flash media works specially on our own device. Glacier/MT4G is formatted using SLC-EUDA. Which is where the problem lies as you can't actually do low level format to the chip. Which would normally fix the corrupted partition table as the quota for the media would be defined normally by doing BINFS format and re-declaring the cylinders you can fix it. But we can't so we only have access to FS format on already pre-defined partition which is already declared as it can't be done again other than replacing the particular eMMc.
Sent from my HTC Glacier
Are you serious my keyboard freezes all the time!!!!
Sent from my HTC Glacier using xda premium
Sent from my HTC Glacier using xda premium
Our Apps
Get our official app!
The best way to access XDA on your phone
Nav Gestures
Add swipe gestures to any Android
One Handed Mode
Eases uses one hand with your phone