AMD Restructuring: What You Need To Know

[somebodyspecial]

The article forgets to mention they've been restructuring now for 5 years, and have previously laid off 30% of their workforce over said time. So not exactly a drop in the bucket when you consider they KEEP doing it repeatedly. I really hope ZEN is a hit and they chose a DIE size that is equal to Intel's whole cpu/gpu die, RATHER than just matching Intel's cpu side. The whole size would mean victory for a while and pricing power perhaps for a few years, while matching the cpu side just means Intel will price it to death until they come up with something faster again. If ZEN's die is huge (~1/2 of Intel's current dies are GPU), Intel's only choice would be to go 8 core. In a world where 4 faster cores are better than 8 (in most cases, except stuff like ripping etc) for most apps/games, they will end up like AMDs 8 core vs. Intel's 4 right now. That would make them very profitable for a few years as Intel would end up 125w etc vs. AMD's 85 or so (like the reverse is true now). IE, when my Devils canyon is ripping it's hitting 60-68w (gpu off), so double that if Intel's response is two cpus.

If AMD went small or barely bigger than Intel's cpu side, Intel can still cherry pick (ala devils canyon but a bit better of course) and still price them down. AMD needs a large die to come back big. Anything less is failure. Saving a few dollars per die and losing is nothing compared to winning and charging $350-400 instead of $160-200 and being slower in most crap. You need to win, and do it DECISIVELY. I would gladly pay $400 for a die size that is the total size of skylake (or whatever Intel is running then) but far faster at like watts. Lopping off the gpu portion and a die shrink + Finfet should allow AMD to win in EVERYTHING if they chose wisely. If not, I guess I'll be buying Intel again for my dad's PC. I couldn't put off my own pc any longer, too much ripping, but hope to buy AMD this time. Last time they had a king for 3yrs they couldn't produce more than ~20% of the market. That would not be the case this time, as AMD could have multiple people producing chips (GF, TSMC, possibly samsung at some point since they share tech with GF). This is the perfect time to strike as Intel has been concentrating on battling ARM and likely has no HUGE die chip they've been designing (quad I mean) as a backup plan if AMD pulls this. It would take a few years no matter how much money Intel threw at the problem to fix this. I really hope this die is >200mm^2, heck I hope it's near 250 Go big or go home. Intel's dual core broadwell is ~133 IIRC, so ~200+ sounds good. The 9590 is ~320 IIRC, so they've been larger before (yeah different process, but you get the point, go big!).

 

[kalmquist]

The article forgets to mention they've been restructuring now for 5 years, and have previously laid off 30% of their workforce over said time. So not exactly a drop in the bucket when you consider they KEEP doing it repeatedly.

I've got to agree with this. Nvidia spent more that AMD on R&D for the first time in 2013, and AMD now has less that 20% of the discrete graphics market. It's easy to claim (as the article does) that "the restructuring will more than pay for itself in roughly a year's time" by looking only at the cost side of the picture, and ignore the effect on sales.

 

[wtfxxxgp]

I doubt 500 workers will be taking it with a grain of salt or if they care whether its a small or large amount compared to other companies.

I don't think this sentence was a fair one. The purpose of the article is not about human tragedy - it is purely about the goings-on at the enterprise-level. We all agree that it is irrelevant whether one person loses their job or 50000 - it is sad news for whomever it happens to. This is not a socialist site - it is about information, not feelings.

 

[uglyduckling81]

I doubt 500 workers will be taking it with a grain of salt or if they care whether its a small or large amount compared to other companies.

I don't think this sentence was a fair one. The purpose of the article is not about human tragedy - it is purely about the goings-on at the enterprise-level. We all agree that it is irrelevant whether one person loses their job or 50000 - it is sad news for whomever it happens to. This is not a socialist site - it is about information, not feelings.

I understand that but the article goes from talking about laying off a bunch of workers and then says we need to take it with a grain of salt. It implied to me that the laying off of so many people is meaningless which it is not.

I know it's not a human tradgedy story but it still seemed a poor way of getting a point across. Could of just said 500 workers have been laid off and left the grain of salt part out.

 

[wtfxxxgp]

I doubt 500 workers will be taking it with a grain of salt or if they care whether its a small or large amount compared to other companies.

I don't think this sentence was a fair one. The purpose of the article is not about human tragedy - it is purely about the goings-on at the enterprise-level. We all agree that it is irrelevant whether one person loses their job or 50000 - it is sad news for whomever it happens to. This is not a socialist site - it is about information, not feelings.

I understand that but the article goes from talking about laying off a bunch of workers and then says we need to take it with a grain of salt. It implied to me that the laying off of so many people is meaningless which it is not.

I know it's not a human tradgedy story but it still seemed a poor way of getting a point across. Could of just said 500 workers have been laid off and left the grain of salt part out.

I hear you, but I still think you're taking the grain of salt bit out of context - it's not about the people, but rather about the entity. I was retrenched once, so I do understand where you're coming from. I just didn't think that the writer was trying to make it seem as though the loss of jobs was not an issue - in the greater scheme of things, when we talk efficiency, we sometimes forget that it has a possible cost: jobs being made redundant. It's not only when you're doing poorly that jobs could be lost - if you're doing exceedingly well it could result in precisely the same thing, unless the expansion planning and workforce planning has been done brilliantly (so, efficiency doesn't ALWAYS cost jobs, but it could).

 

[Tzn]

Fix your god damn power-consumption/performance balance AMD, that's all that you need to do to be a strong contender again and that's the main problem with AMD.

 

[velocityg4]

Of course management will still get all their bonuses. Even if the company is losing money.

 

[kiniku]

No one believes AMD is going to go under as soon as next week. But OP you were provided a statement that was probably written initially for shareholders. If you can't see the obvious that most the readers here can see I'll clue you in, AMD is in trouble.

"AMD was in talks to sell roughly 25% of itself to Silver Lake Management, but that negotiations are currently on-hold due to differences in strategy and overall cost, Bloomberg reports. AMD has good reason to push for as high a price as it can get, seeing as the company’s future largely depends on securing a lifeline until the launch of Zen, its next-generation CPU architecture that’s not expected until 2016. AMD wasn’t willing to comment on the current specifics of its situation, and neither was Silver Lake, for obvious reasons."

http://www.extremetech.com/computing/215353-amd-talks-with-private-equity-firm-silver-lake-fell-through-report

 

[Larry Litmanen]

AMD put too much effort in making APU's but they didn't put any effort on making general CPU's without powerful GPU's. They could've make cheap 8 core cpu's or power powerful cpu's that could've competed with Intel I3/i5/i7

That's true for gamers, but majority of sales don't come from gamers but rather from the general public so you have to keep that in mind.

For a general public a CPU with a nice GPU is a great solution.

 

[Chris_A79]

I think the restructure is a good move. I'm not party to the internal politics in their business but it's not enjoyable to be constantly second fiddle to another business in your sector - some folks in AMD will be out their to win!

Restructuring can bring focus to essential strategic targets and objectives but it easy to get hideously wrong if you end up with the wrong people in the wrong places.

For me, AMD have an opportunity for a market that creates a balanced system. The right number of cores, the right power GPU and a fat enough bus. I concur with a prior poster about multiple sockets for a scalable system, especially if you can demonstrate a fully stacked system has some real power.

Maybe it's time that AMD joined forces with another company for a rebrand. LG or Samsung APU perhaps? I think for too long the brand name has meant inferior to Intel. Their products are more than competitive for most users but they are unable to capture that market space.

 

[jimmysmitty]

I dont blame their engineers as much as how many they have employed is minuscule compared to Intel and Nvidia.

AMD actually has more employees than nVidia, much less than Intel. Of course nVidias current 8800 are all focused on just graphics but AMD was well aware of this when taking on ATI. They knew they would not be able to give the same focus nVidia has.

Although ATI did very well and probably had less employees than nVidia did so amount of employees or engineers is not always what determines if it will be better or worse. Remember the 9700Pro? Came from far left field to smack nVidia upside the head. Or Athlon 64, of course that was a two fold where AMD made a good chip and Intel made a bad one making K8 look even better.

Personally I feel bad for the IT people and the employees who will now have to rely on external IT support. The external IT people don't tend to get paid as much as internal, entry level is $12-15 bucks an hour, and they don;t always have the best quality of people, something AMD will now have no control over.

I guess this is better than them just sitting and staying the course. Hopefully it has a good outcome in the end as I have been feeling that AMD has been lacking in most of their markets recently unless you are on an extremely small budget. Their GPUs were doing fine for a while but it is really hard to suggest a Fury X when a 980Ti does the same performance using less power and not requiring water cooling to do so.

Only the future will tell. Either we will reminisce about the good old days of AMD or we will have good competition spurring faster innovation again.

 

[James Mason]

The internal IT support is just things like email, logins, domains, and physical computer support. (gonna be "hard" to outsource that maybe, but that is my exact job, an "outsourced" internal physical computer support for an organization. My company is hired by another company who was contracted by the place I actually work at.) Losing those employee's won't affect AMD's product quality.

But I'm still worried that AMD is gonna die. I think if they spin off the graphics division into another company, it could survive on it's own, but AMD's CPU business has just not been innovative enough to compete with Intel at all.

Becoming a monopoly because your all your competitors have died out will be interesting to see how the FCC handles something like that.

 

[Chris Droste]

i honestly think if they could just harness HBM into a die shrink (at this stage even a single shrink to 22nm would help massively) maybe 16nm? and build a CPU that's just simply an 8370 that's been refined again to incorporate HBM, i could declare them back in the game. they're hitting the limits on their parallelism philosophy when they got away from the Phenom 1090T/1100 era that still to this day can stand up to their most recent chips when the benchmarks are out. right now they need to stop adding "virtual" or real cores, and prep strictly for IPC on a new die process, and they need it last year. heck; introduce a few "Fury lite" graphics to bolster their bottom line so they can get a contract with a fab. an FX9590 with a sub-100w power envelope would be music to the ears of enthusiasts and enterprise users alike, roll out HBM-based CPUs on 22 and 16nm. do it do it doit and stop waiting for the enterprise sector to pick up on almost 3year-old "value priced" cpus

 

[ohim]

Only way out is for Intel to let go to the x86 licence to the company that could buy AMD (AKA Samsung) but at this moment if this ever happens , the x86 license won`t go to Samsung. Intel already hates competition, they proved this back in the Athlon 64 days.

 

[xenol]

The biggest issue I feel that AMD missed is that the push in the last few years has been towards better efficiency, rather than raw performance. Raw performance is great and all, but over time if the cost to run the hardware is significantly greater than its competitor, even if its competitor takes a bit longer to do the job, then they're going to go with the competitor.

And there's one other sector that AMD has failed miserably in that Intel has enjoyed practically unprecedented: laptops. Sure, AMD has GPUs in there, but most people aren't buying laptops with discrete GPUs. And sure, AMD has APUs in laptops that offer better graphical capabilities than Intel's current offerings, but most people don't care about that either. In the end, if an AMD laptop has poopy battery life, it's a foregone conclusion in laptops.

 

[littleleo]

AMD's moves over the last few years is what has put them in this position. With this reorganization I hope they are not just moving the chairs around on the Titanic. So many errors the latest is HDMI 1.4 connector on the Furys, Bad crossfire drives for the Furys. Putting all their eggs in the APU basket which is a product with limited appeal and without a clearly defined market. Abandoning the gaming CPU segment completely... I pray the Zen is a winner it needs to be at least equal to the Skylake. I got to wonder with all the layoffs to their R&D if they have enough talent left to pull this off. AMD & Intel hate each other Intel will never help AMD. Cross your fingers boys and girls and pray to the Multi-Core God above to help them make this work.

 

[alextheblue]

It is really hard to suggest a Fury X when a 980Ti does the same performance using less power and not requiring water cooling to do so.

Though far from perfect, their new cards are a big step in the right direction. Fury X uses a little more power than 980 Ti, but they really closed the gap in terms of power and performance compared to Hawaii. They need to continue in that direction if they want to stay competitive.

Oh and minor point: just because it comes with a CLC doesn't mean Fury X "requires" a water cooler. They could have chosen any number of aircooling configurations. They decided to go premium on the X model. Look a the Tri-X Fury as an example of an aircooler that is way more than enough, and actually provides a lot of headroom.

 

[photonboy]

CPU vs GPU irony:

When AMD designed CPU's like the FX-8350 they assumed the software would soon come for games and application to utilize all those "cores" effectively.

So they basically bet on having MORE cores that individually performed worse. Unfortunately that didn't pan out.

GPU:
Now AMD did actually think ahead with some of their GPU Compute architecture. Keep in mind the "new" R9-390 is a rebadged R9-2xx which is basically a rebadged HD7950/70 (don't know exact models).

My point is that people are suddenly looking at benchmarks like AoS where AMD GPU's can make use of certain new features to pull ahead of NVidia. Whether it's going to pan out isn't my main point but rather THIS:

*AMD pushed a GPU architecture that, just like their CPU architecture, was too forward thinking at the expensive of the present.

(On a side note people are now critisizing NVidia's GPU architecture... people, they optimized for the games that were out NOW not some future games. Even if you bought a game today that doesn't mean that more than 50% of the games in the next year will take advantage of ASync Compute or whatever... plus the poor DX11 drivers that never really got optimized since they don't seem to have the resources.)

Summary:
Anyway, I'm no expert but I see part of AMD's problems stemming from poorly understanding how things would change in both the long and short term. Again, put another way, optimizing to the FUTURE at the expense of the PRESENT.

That then may have affected sales and set about a downward spiral due to lack of funds for RESEARCH which then means selling same CPU's and GPU's... not sure how you claw back from that since investments take YEARS to pay off.

 

[Afrospinach]

Who knows, Intel may bail them out before long. It's probably in their best interest that AMD survive, lest they go under and suddenly Intel is a monopoly.

I am not sure why this is being downvoted so much. Microsoft was one of Apple's backers in their lowest of lows.

Becoming a monopoly because your all your competitors have died out will be interesting to see how the FCC handles something like that.

They will have to make all of intel's CPU related patents FRAND so anyone can produce a capable CPU. They cannot exactly cut up a company like that since it's core is knowledge. FRAND = free, reasonable and non discriminatory, a state reserved for patents so crucial to an industry they are mandatory to compete, so in short they are somewhat set up for this eventuality.

Maybe it's time that AMD joined forces with another company for a rebrand. LG or Samsung APU perhaps? I think for too long the brand name has meant inferior to Intel. Their products are more than competitive for most users but they are unable to capture that market space.

I can never find the details when I want to, but intel has some interesting licensing deals with AMD's x64 that makes their CPU business a pretty unattractive buy(something like any new owner will NOT be able to use x64 without making a new agreement first since it is based on intel x86). It is likely AMD is very restricted in such activities.

 

[jaber2]

AMD, you need to justify your existence

 

[jimmysmitty]

It is really hard to suggest a Fury X when a 980Ti does the same performance using less power and not requiring water cooling to do so.

Though far from perfect, their new cards are a big step in the right direction. Fury X uses a little more power than 980 Ti, but they really closed the gap in terms of power and performance compared to Hawaii. They need to continue in that direction if they want to stay competitive.

Oh and minor point: just because it comes with a CLC doesn't mean Fury X "requires" a water cooler. They could have chosen any number of aircooling configurations. They decided to go premium on the X model. Look a the Tri-X Fury as an example of an aircooler that is way more than enough, and actually provides a lot of headroom.

To a point. We won't know if Fury X needs a CLC or not as the closest to it is Nano but nano is limited in TDP and due to that does not perform the same.

There is almost no OC headroom on the Fury core itself though as clocks seem to top out around 1100MHz for most. Of course it seems we will never really know since Fury X seems to be a non customize-able part by OEMs or else we would see some by now. We do know though that a 980Ti can go well beyond its default clocks, the Asus Strix can push over 1300MHz base with a boost into 1400MHz. I sort of wish AMD left the designs for cooling up to the OEMs as they seem to come out with way better designs than AMD or nVidia do.

 

[jerryman]

I am very happy that there is not just one CPU GPU maker's that why I buy AMD.
Wish AMD good luck!

 

[AMDHTPC]

I "need" to know about AMD's restructuring? Why, exactly? None of my desktop PCs in the last 20+ years have contained an AMD chip, nor will they ever in future.

I guess you want an Intel good customer cookie then?

Please tell us some more neat info we care absolutely nothing about.

 

[the1reddog]

Just curious if youve seen orbheard of the device that turns any flatscreen tv into a computer. With up to 8 gs of ram and supposedly graphics to impress any gamer, the whole thing is the size of a flashdrive and plugs into hdmi socket on your tv. For me, a guy who just built my first 2 gaming omputers, this tech both cascinates me, and pisses me off, lol. I just heard the device will be out dec this year
...

 

[Chris Droste]

those are Compute sticks, and their graphics are still far inferior to your gaming builds. some of the more recent ones with cherry trail or maybe even skylake MAY have enough graphical horsepower to play some older games but i wouldn't bother asking them to play anything within the last 5 years. if you wanna play Quake3 and Counterstrike, maybe some Battlefield 2/3? it could pull it off but to what end?
I'm waiting for when AMD can put some R9 graphics on an APU; that would make me happy. HBM/Kaverni/380 graphics? okay NOW I'm sold on an APU.