|
SourKraut posted:Sir, this is America. No such thing exists. To be fair, my first level manager is loving awesome. He’s been with our team like 7-8 years now and just gets all of us. I’m on a very high-performing team and he knows we get our poo poo done, and done very well. If we just suddenly take a day or two off to go skiing or go to a kid soccer tournament or whatever, he doesn’t care. He knows we’ll go above and beyond to get our poo poo done. He also recognizes work/life balance and is completely understanding that when 5:00 rolls around, I’m done caring about work until 8am the next morning. Never tries to get us to work extended hours. He even schedules happy hours on Fridays every now and then and will buy the first round. It’s very refreshing, and he’s probably only the second manager I’ve actually liked in my almost 2 decades here. I’ve had some really bad ones who have tried to gently caress me on my yearly review after having a good year. I consider him the exception to the rule, and honestly part of the reason I’ve stuck through subpar raises for a while. Henrik Zetterberg fucked around with this message at 06:00 on Feb 6, 2023 |
#
?
Feb 6, 2023 03:22
|
|
|
|
| # ? May 18, 2024 02:39 |
|
|
Most of my direct managers were good -great, the only problem was that it was somebody new every year or two. I started in the fab but moved out a few years ago and that helped , the real gently caress ups at intel are the senior area managers and vps who despite repeated gently caress ups are still mostly in charge It took way too long to soft fire peng (he wound up jumping ship after an annoying stint in supply chain and works for a Chinese firm now), firing Myra took way too long, sohail had to be caught lying to get canned, the list goes on
|
|
#
?
Feb 6, 2023 03:47
|
|
|
Intel senior managers still feel like they need to micromanage everything, I know vp level folks who are still doing loving lot dispo, it’s ridiculous
|
|
#
?
Feb 6, 2023 03:59
|
|
|
I had a department head come in, tell everyone we were spending too much time and money doing validation and that he was canning our validation tool suite and cutting all the extraneous validation work we were doing and kept harping on “use case” Then a customer bitched him out in a call for the number of bugs the product had, referred to it as an unbaked pile of sand, and he sent out an angry “no, not like that” email to us He beat out my old boss who I liked for the position and the first thing he did was unmanagerize him and replaced him with a parade of people who retired/got fired after 6 months WhyteRyce fucked around with this message at 04:11 on Feb 6, 2023 |
|
#
?
Feb 6, 2023 04:06
|
|
|
WhyteRyce posted:an unbaked pile of sand, Now there’s a line to save for a rainy day…
|
|
#
?
Feb 6, 2023 04:24
|
|
|
With all the poo poo that is going on it’s not too surprising we can’t get any support from intel for our SPR pdks, we have been been blocked for ages since the ITP probes stopped working for us. Rolling back to an older bios works at least but wtf dudes.
|
|
#
?
Feb 6, 2023 04:39
|
|
|
wet_goods posted:Intel senior managers still feel like they need to micromanage everything, I know vp level folks who are still doing loving lot dispo, it’s ridiculous Yeah I work with one who does exactly that, and I can't believe I didn't think it was weird until you put it that way
|
|
#
?
Feb 6, 2023 07:14
|
|
|
What's a lot dispo and why is it weird?
|
|
#
?
Feb 6, 2023 08:44
|
|
|
WhyteRyce posted:I had a department head come in, tell everyone we were spending too much time and money doing validation and that he was canning our validation tool suite and cutting all the extraneous validation work we were doing and kept harping on “use case” LOL, what a tool. And what an indictment of Intel's management culture that (a) he got put in that position and (b) didn't get taken out of it as soon as he proposed something as dumb as "slash validation to the bone". I remember stories about Apple being very displeased with Intel releasing buggy silicon to manufacturing starting some time around the Skylake generation. Related?
|
|
#
?
Feb 6, 2023 08:48
|
|
|
WhyteRyce posted:I had a department head come in, tell everyone we were spending too much time and money doing validation and that he was canning our validation tool suite and cutting all the extraneous validation work we were doing and kept harping on “use case” No, not like that! You're supposed to ship a perfect product without validation, why were mistakes made? Testing isn't necessary if nobody screws up!!
|
|
#
?
Feb 6, 2023 10:27
|
|
|
Is this the hardware equivalent of "I test in prod  "? We started getting free therapist appointments this year but they're always booked the moment any slots become available so that's one more thing to be depressed about
|
|
#
?
Feb 6, 2023 10:44
|
|
|
mmkay posted:What's a lot dispo and why is it weird? Making individual decisions on how to proceed with a lot (one FOUP or containers worth of a handful of wafers) when something comes up. Usually it’s some kind of misprocess that could impact the yield of that lot.
|
|
#
?
Feb 6, 2023 15:46
|
|
|
Yeah... hardware is kinda the opposite of software in theory but given how much that used to be done in software has moved into hardware more software-like techniques probably need to be incorporated into silicon anyway. See: https://www.usenix.org/conference/osdi21/presentation/fri-keynote
|
|
#
?
Feb 6, 2023 15:51
|
|
|
necrobobsledder posted:Yeah... hardware is kinda the opposite of software in theory but given how much that used to be done in software has moved into hardware more software-like techniques probably need to be incorporated into silicon anyway. See: https://www.usenix.org/conference/osdi21/presentation/fri-keynote Just buy an FPGA company!
|
|
#
?
Feb 6, 2023 15:56
|
|
|
BobHoward posted:LOL, what a tool. And what an indictment of Intel's management culture that (a) he got put in that position and (b) didn't get taken out of it as soon as he proposed something as dumb as "slash validation to the bone". No was a different, major customer with this specific quote I remember a few months before this incident he set up an uncatered lunch with a small group of us. Was supposed to be an informal temperature check for the department. When we talked about how poorly we felt things were going he feigned interest in helping. But when we kept pressing him on time and resources he eventually curtly told us “guys you’re not listening to me, I’m giving you the support and approval to cut whatever you need so we can get this out on time” His angry email later said there was a misunderstanding with his message and he never said to ship something out that wasn't ready or had bugs WhyteRyce fucked around with this message at 17:59 on Feb 6, 2023 |
|
#
?
Feb 6, 2023 16:58
|
|
|
wet_goods posted:Making individual decisions on how to proceed with a lot (one FOUP or containers worth of a handful of wafers) when something comes up. Usually it’s some kind of misprocess that could impact the yield of that lot. To add, it's something most lowbie engineers should be able to handle on their own (unless they're truly useless). Or even technicians depending on the area. I've never worked in the main fab, but if I had to guess it's probably folks trying to protect themselves from blame by having other modules take the hit of a wafer scrap or something stupid like that. There can be a lot of infighting and blame-shifting due to how tool ownership is split in the fab, so you spend a lot of time making sure nobody points the finger at your tool for being the issue (even if it is!). So I could see there being a pissing war getting micro-manager VPs involved. At least in my case it's not really about that, it's more about which wafers get split out for post-Si testing. But there's still a difference between signing off splits and actually dispositioning them yourself.
|
|
#
?
Feb 6, 2023 17:44
|
|
|
haha the Folsom open forum video was taken down, probably because Pat was talking too off the cuff, so now Blind gossip about what was said becomes the source of truthHalloKitty posted:No, not like that! You're supposed to ship a perfect product without validation, why were mistakes made? Testing isn't necessary if nobody screws up!! they at least tried to be slick about it and said things like "this is a legacy IP block with no changes why do we need to test it". Unfortunately telling them that this legacy IP block talks to a definitely not legacy block or is used as part of a definitely not legacy flow generally fell on deaf ears. Also synthetic testing is bad. Don't do it. You're creating these wild exotic scenarios that won't ever happen in the real world. Customers won't see it so why bother...hey why are these customers seeing all these bugs now is 3dmark not sufficient to hit this? WhyteRyce fucked around with this message at 18:38 on Feb 6, 2023 |
|
#
?
Feb 6, 2023 18:33
|
|
|
Any of you Intel folks hear about a fire in a lab a couple weeks back? Wherever the CXL testing is going on (involved a competitor of my co, lol) Listening to a DDR5 PMIC when it is screaming about temperature warnings is probably a good idea.
|
|
#
?
Feb 6, 2023 18:48
|
|
|
WhyteRyce posted:haha the Folsom open forum video was taken down, probably because Pat was talking too off the cuff, so now Blind gossip about what was said becomes the source of truth That’s not surprising. I watched most of it, and there were a few comments that made me snap back to the screen and made me think “holy crap, he said what?!” It is slightly refreshing to have a CEO actually speak his mind from time to time. I’m guessing they edit the video and re-upload it?
|
|
#
?
Feb 6, 2023 19:12
|
|
|
WhyteRyce posted:I had a department head come in, tell everyone we were spending too much time and money doing validation and that he was canning our validation tool suite and cutting all the extraneous validation work we were doing and kept harping on “use case” Intel: An Unbaked Pile of Sand
|
|
#
?
Feb 6, 2023 19:57
|
|
|
WhyteRyce posted:haha the Folsom open forum video was taken down, probably because Pat was talking too off the cuff, so now Blind gossip about what was said becomes the source of truth Definitely rings of other times blocks of code (or in this case, a piece of hardware) were used in a new product without thinking it had any flaws because hey, it worked before, without taking in to account the new ways it could cause problems in another system
|
|
#
?
Feb 6, 2023 20:13
|
|
|
FuturePastNow posted:Intel: An Unbaked Pile of Sand What is Intel? A miserable little pile of unbaked sand. But enough talk, have at you! <gives employees a 15% salary haircut>
|
|
#
?
Feb 6, 2023 21:37
|
|
|
Henrik Zetterberg posted:It is slightly refreshing to have a CEO actually speak his mind from time to time. https://youtu.be/AA1Q_MhYIjo
|
|
#
?
Feb 7, 2023 02:10
|
|
|
WhyteRyce posted:I just remember Murthy and he’s probably thrilled that everyone forgot how much they hated him and how much a failure he was Murthy took home over $100M in compensation that would have been better spent on a litho tool. I saw him do a campus visit one time and he had an entourage of ~10 people surrounding him, including two security people. It was really weird because I would often see the CEO and other very senior people walking around the building by themselves just like everyone else. When BK left and the CEO position became open, he did a self-promotion tour across multiple sites holding "open forums" where he shared his story of overcoming adversity. He was the "awkward, chubby kid, bad at sports, from a midsized city in India" who rose to great heights in engineering!  (Also, his parents were loaded old-money 1%'ers and sent him to elite boarding schools and prestigious foreign universities)wet_goods posted:It took way too long to soft fire peng (he wound up jumping ship after an annoying stint in supply chain and works for a Chinese firm now), firing Myra took way too long, sohail had to be caught lying to get canned, the list goes on  Sohail was such a bully, and they waited ~5 years too long to replace him. Stacey would have been miles better as CEO than BK. I really liked and respected my direct managers and +1's for ~90% of my time there. The boneheads get very deeply entrenched at the exec and exec minus 1 levels.
|
|
#
?
Feb 7, 2023 05:13
|
|
|
Murthy: “I’m going to coin the term 5G fog and bring it up everyone time I speak. Also would you like to hear about rugby and how I played it”
|
|
#
?
Feb 7, 2023 05:16
|
|
|
Murthy was so hilariously bad and his compensation to performance ratio has to easily be the worst in company history.
|
|
#
?
Feb 7, 2023 05:48
|
|
|
WhyteRyce posted:Murthy: “I’m going to coin the term 5G fog and bring it up everyone time I speak. Also would you like to hear about rugby and how I played it” 5G and really wireless in general is a contest to coin terms and it’s seems to sillier the better. 5G fog computing has to be close to the top though.
|
|
#
?
Feb 7, 2023 07:49
|
|
|
hobbesmaster posted:5G and really wireless in general is a contest to coin terms and it’s seems to sillier the better. I remember the week he was fired he had some bullshit six pillars company wide meeting planned and it was the funniest poo poo ever to see the invitation to that and the notice that he got canned in my inbox
|
|
#
?
Feb 7, 2023 07:58
|
|
|
Some other funny and weird corp stories and gossip. Aircraft: For context, Intel has a leased set of aircraft that run fixed routes between the major campuses (AZ, Folsom, Santa Clara, and Oregon). They're "free" to use for business travel, and typically cost less than flying commercial. Everybody uses them, from the summer interns up through the C-suite, and because you fly from the exec terminals you don't go through all the TSA/security nonsense and airport rigamorole. The flights nearly always leave without empty seats, so if you can't reserve an air shuttle seat you probably end up flying commercial. When Bob became CEO, he insisted that they lease a private jet for him to use so he didn't have to get on the shuttle with the rest of the rabble like us. One dude got fired for abusing the air shuttle, because he was taking several flights a week on it. He was based at one site and a new role relocated him to another site at great expense. He took the relo package (probably $10k+) and then just didn't move and instead commuted via air shuttle and got a small apartment. He did this for almost a year, so those 100+ flights he took pretty much directly translated into other people buying 100 commercial flights  . This guy had been working there over 10 years, and it caused some embarrassment to his spouse who also worked there in a pretty senior role.During the Cougar Point disaster with the catastrophic SATA bug on the PCH, air shuttle service was suspended because the planes became improvised cargo planes. They pulled out the seats, filled the planes in AZ with fully loaded FOSBs of the finished, corrected wafers, and flew them day and night up to Oregon to the die prep facility. The people who lived by the airports were pissed, because there usually weren't big planes like this taking off and landing at these small regional airports at 3 AM. Brian: BK has met all of his spouses and romantic partners at work, all of them process engineers. Seems he has a type. In 2011, BK was in Arizona buying a car for his daughter. The sales guy was chatting with him, as they do, and BK told him he worked for Intel (though no indication that he was running the ~40k employee manufacturing org). The sales guy said, oh, that's so cool about the new $4B factory being built, my cousin is in construction and is going to be working on it. The factory had not yet been announced and was supposed to be under strict NDA, and he was furious. After the news broke about BK planning a Trump fundraiser at his home in 2016 (which he cancelled), the company spent a 7 figure amount on adding security infrastructure and personnel to his house. Heard this one second-hand from someone who was told by one of the people present. ~15 years ago when BK was in some manufacturing leadership role long before he was CEO, he was meeting with a senior PE to tell him that they needed to lay off his entire team because of scheduling pushes. The PE pushed back and said that's stupid, because you're going to want these people back in 8 months and they have a pretty unique skillset, I'm not going to go along with this. It continued to heat up and BK took a swing at the guy. They both landed a few punches before the other people in the room pulled them apart, nobody got fired, and life went on. Factory: A factory appealed their headcount targets for the next year asking for more staffing. This was based on an analysis they had done (how?) estimating how many factory employees might become pregnant and would thus need alternate work arrangements. Night shift in one of the factories had an open secret amongst the techs of a hidden nap space. Pull up one of the raised floor tiles and someone had made a little bed in the subfloor with a mattress and pillow made out of cleanroom grade anti-static bags. During a factory conversion, they'd find contraband stashed in the darkest corners of the cleanroom. Empty liquor bottles and candy wrappers, but the weirdest was a bucket of KFC chicken, complete with discarded bones. A contractor servicing the fire suppression system in the cleanroom did not correctly depressurize the sprinkler line, and ended up accidentally turning on the deluge of sprinklers right above ~$50M worth of process-limiting machines. There was a fire in the factory once, and the entire facility was evacuated for over 2 hours. Fire suppression did it's thing and nobody was hurt, and the factory continued to run autonomously the whole time without any loss of product. Weird money: I was once on an email chain with maybe 15 replies in under an hour that had some big names on it, including someone two levels down from the CEO. I was worried it might be something urgent or important, but it was about which department to expense ~$600 of cat food towards so they can feed the feral cats onsite. Someone managing a pretty big organization of ~500 people got fired after repeated warnings for stealing snacks from the cafeteria. They were probably clearing $200k+ a year and got canned for petty shoplifting of hospital cafeteria-tier food. Once saw a recurring $200/mo charge to a defunct facility for telecommunications service. Upon investigation, it was a recurring billing for a 10mbit DSL line going to a law office offsite. The law firm had been engaged for some patent work in 1998 and provided that expensive $200/mo DSL line. They finished their work within a year or so, but the company kept paying the internet service bill for 20 more years. The law firm had actually moved ~15 years earlier, so it was providing expensive and slow internet to an empty office owned by someone else. A department of 200 people had 600 mobile phone lines billed to them monthly. Some were for phones they no longer had, others for data-only SIMs they had sitting in their desks that they never even used on a tablet or hotspot. Lots of fraud stories of stupid people being caught in stupid ways. One was a conspiracy to hire several "ghost" employees and collect the salaries. Another was expensing a sports car for the purpose of a "mobile communications test suite", and the test equipment was purchased but never installed because it didn't actually fit. Sales guy expensing his entire family plan's 6 mobile phone lines as a work expense. Sales guy putting a $2,000 set of tires for his vehicle on the corporate card.
|
|
#
?
Feb 7, 2023 22:03
|
|
|
drat those are some great stories. The one about the DSL lease for 20+ years made me lol especially.
|
|
#
?
Feb 7, 2023 22:17
|
|
|
I’m either lucky or unlucky I haven’t come across anything like that during my tenure  That DSL one is great. And that stuff about BK demanding his own plane seems spot on since Murthy got one to commute from his SD McMansion to Santa Clara or whatever.
|
|
#
?
Feb 7, 2023 22:38
|
|
|
quote:They were probably clearing $200k+ a year and got canned for petty shoplifting of hospital cafeteria-tier food.
|
|
#
?
Feb 7, 2023 22:48
|
|
|
canyoneer posted:
BK being a huge horn dog was an open secret. My favorite BK Blind rumor was that what got him fired was that his wife found out and barged into an afternoon meeting to tear him apart and the board couldn’t cover for him anymore
|
|
#
?
Feb 7, 2023 23:49
|
|
|
necrobobsledder posted:Yeah... hardware is kinda the opposite of software in theory but given how much that used to be done in software has moved into hardware more software-like techniques probably need to be incorporated into silicon anyway. See: https://www.usenix.org/conference/osdi21/presentation/fri-keynote lol yeah, hardware's got a lot to learn from the upstart discipline half as old that churns out pure crap at speed and patches it in the field WhyteRyce posted:I had a department head come in, tell everyone we were spending too much time and money doing validation and that he was canning our validation tool suite and cutting all the extraneous validation work we were doing and kept harping on “use case” I remember a project came through, some ex-AMD folks who were convinced they had the secret to Doing Chips and had some stolen Native American name. Talking up all the great architectural improvements, all the fantastic design work they'd enable. Just a greenfield project where you could attract all the heavy hitters and they'd crank it out. "Validation? It's a solved science. We really don't want those rockstars, if they even exist. Validation is boring execution and we'll get the minimum count of warm bodies to fill that, I guess." Now, I don't have the direct knowledge that's why they hosed up? But it certainly didn't attract anyone from my team to give them the time of day.
|
|
#
?
Feb 8, 2023 00:11
|
|
|
Validation crew 
|
|
#
?
Feb 8, 2023 00:55
|
|
|
canyoneer posted:Another was expensing a sports car for the purpose of a "mobile communications test suite", and the test equipment was purchased but never installed because it didn't actually fit. Such a waste, the guys that tear apart and rewire the demonstrator cars at my company have a blast. You gotta get a max trim Escalade, navigator or wagoneer or something (they’re previous model year reviewer cars iirc)
|
|
#
?
Feb 8, 2023 01:15
|
|
|
canyoneer posted:the Cougar Point disaster Please mods I need this username
|
|
#
?
Feb 8, 2023 05:53
|
|
|
necrobobsledder posted:Yeah... hardware is kinda the opposite of software in theory but given how much that used to be done in software has moved into hardware more software-like techniques probably need to be incorporated into silicon anyway. See: https://www.usenix.org/conference/osdi21/presentation/fri-keynote I listened to this talk just now and it doesn't support your idea, though I can see why you might have misinterpreted it that way. From the perspective of a hardware designer, the topic of his speech is more like "what used to be done in software is still done with software". What he's annoyed about is that it's moved to a software layer he cannot control, inspect, debug, or alter. Basically, in any modern SoC, there's a shitload of tiny assistant CPUs. They run firmware implementing functionality which, in the classic abstract machine model he's ranting about, would either be hardware state machines, or part of device drivers and hence actually part of the OS. So now the relevant questions become: Why not hardware state machines? Why not ordinary device drivers? The answer to the first is that if we RTL engineers actually put in a shitload of complex hardware state machines, we'd have to debug them before tapeout. It is insanely hard to do this kind of work in simulation, so the bugs would not get found until validation, and we'd have to do a bunch of chip spins to fix them. FPGA and other forms of pre-silicon emulation can help, but probably not enough. The answer to the second is that often, you need to make control or other decisions quite frequently, but it would be wasteful to wake up a real CPU core for every one because the total compute power required is only a tiny fraction of a percent of a real CPU. Also, these decisions often come with hard realtime requirements, which are difficult to provide on cores shared with user applications. A simple 32-bit or 64-bit RISC core with fully deterministic timing costs an inconsequential amount of die area these days, so the easy path is to just slap one down and clock it so low it uses nanowatts. And you don't put down just one. They're so small, you just put in one per subsystem which needs it. IIRC, there's something like 20 "Chinook" microcontroller cores in every Apple M1 SoC.
|
|
#
?
Feb 8, 2023 10:40
|
|
|
JawnV6 posted:lol yeah, hardware's got a lot to learn from the upstart discipline half as old that churns out pure crap at speed and patches it in the field I don't know how many times I've had to hear designers complaining about validation people just coming up with these implausible, unrealistic scenarios that waste everyone's time when it's all emulation or A0 testing and then immediately start bitching about coverage gaps when it it's a high priority customer issue blocking launch Those complaints about validation wasting time and resources got more traction when PC sales start to slip and managers get desperate to find some way to maintain profit margins above all else. WhyteRyce fucked around with this message at 19:07 on Feb 8, 2023 |
|
#
?
Feb 8, 2023 19:01
|
|
|
|
| # ? May 18, 2024 02:39 |
|
|
Am I correct in quoting (likely from someone in this thread) that the first few seconds of first power-on of Nehalem executed more clock cycles / instructions than had ever been simulated for the design? Or was that my old boss (ex-Intel) exaggerating slightly? I’ve done logic design / validation before, but nothing at the scale of a modern CPU or experience with leading-edge validation tools, so I’m curious.
|
|
#
?
Feb 8, 2023 20:59
|
|
