I'll start this discussion off by linking this article, written by Anand La Shrimi back in 2006:
https://www.anandtech.com/show/2055
A lot has happened since that acquisition, with arguably the most significant thing being AMD going fabless and spinning off their fabs into a new company, Global Foundries in 2009.
One thing that AMD has had to deal with is the amount of long term debt that the company has accrued. A huge chunk of that debt appeared with the ATI acquisition, and AMD has finally managed to reduce it's long term debt from a high of $5.28 Billion back in 2009 to just over $1 Billion today. The fab spinoff did net AMD a nice chunk of cash to reduce their debt load back in 2009-2010 time frame, but the long term debt continued to hover at or just over $2 Billion for a number of years.
https://www.macrotrends.net/stocks/charts/AMD/amd/debt-equity-ratio
Smarter people than I can comment on AMD's numbers over the last five years. Here's a relevant link for reference:
https://www.marketwatch.com/investing/stock/amd/financials/balance-sheet
One thing that is clear to me, things were looking rather grim around 2015...
Anyways, I did see this comment in the Anandtech comments from the 2006 article, which wasn't that uncommon of a sentiment back in the ATI acquisition days:
s1wheel4 - Wednesday, August 02, 2006 - link
This will be the end of AMD and ATI as we know them today....and the end of both in the high end enthusiasts market...when merged; the new company will be nothing more than a mediocre company both of which will lag behind Intel and NVIDIA in performance.
Looking at the 2015 numbers, this prediction essentially came true.
IMHO, if it hadn't been for the Playstation and XBox wins, well AMD might not be where they are today cash wise and market position wise...
In the last four years, though, we've witnessed a rather incredible comeback on AMD's part. GPUs are still struggling a bit, but Zen has been nothing short of awesome. Sure, Intel is slightly better at gaming and such, but Zen and Intel chips are essentially trading blows. And, for now, well EPYC Rome is quite simply impressive.
Whether AMD can field a high end GPU to challenge Nvidia for the top end crown in the near future still remains to be seen, as Nvidia isn't standing still. Neither is Intel, but right now Intel has a real fight on their hands, for the first time in years.
As for the wafer supply agreement with GloFo, AMD's requirements in this regard were relaxed recently:
https://www.anandtech.com/show/13915/amd-amends-agreement-with-globalfoudries-set-to-buy-wafers-till-2021
This agreement has been a significant expense to AMD in past years, but this latest version of the agreement, while it does carry over to 2024, seems to look much better from AMD's standpoint, at least from what I'm seeing. The fact that GloFo gave up on 7nm for now no doubt played into this amendment.
AMD is still below 20% in both the CPU and GPU markets in market share, but it has been gaining recently. 50% parity is still a long ways off, but we've seen some interesting blips from a few retailers such as Mindfactory in the last month or so on the CPU side. The GPU side is still a pretty tough fight, though. And the laptop market is a tough one for AMD to crack as well, but they are making a valiant effort. Intel's production shortages have helped a bit here.
Also, AMD was essentially nonexistent in the server market just a couple of years ago. Their current market share is still low-mid single digits, but as Lisa essentially said back when 14nm EPYC was introduced, 'We've got nowhere to go but up'. Based on EPYC Rome's performance and TCO numbers, AMD now stands a good chance at increasing server market share significantly in the coming months.
In any case, I just thought that this would be as good a time as any to look back at AMD's decision to acquire ATI, subsequently spinning off their fabs, and where they are today.
So how do you feel about things now? 13 years after the ATI acquisition?
AMD acquired ATI in 2006 for 5.4 billion dollars. For a while, they were competitive with NVIDIA.
When I started building PCs in 2011, both companies were around 50% market share. This time was awesome because you had real options, and market forces worked.
Unfortunately, this 5.4 billion was difficult for AMD to sustain and it almost killed them as a company. At the time Bulldozer was not competitive with Sandy Bridge, and AMD fell further behind in the CPU game.
This meant that driver support got cut to a skeleton crew, and R&D got legitimately slashed.
AMD failed so hard before Ryzen and their stock price got so low that there were legitimate fears that they may fail completely.
Then, among those other problems, NVIDIA used their winning position to get further ahead, and developed new features like DLSS and RT. Suddenly, GPUs took even more R&D because software and hardware both needed way more budget.
I argue that if AMD hadn’t have overextended themselves AND bet on GCN (graphics core net) for FIVE GENERATIONS at the same time that their CPU division was failing, we wouldn’t have a monopoly in the GPU space (nvidia).
If ATI was independent and wasn’t subject to AMDs financial failings and tech blunders, the GPU space would be healthier today. ATI was trading blows with NVIDIA in 2006 and there’s literally no reason to believe they wouldn’t be a serious competitor now. In fact I’d bet my life savings they’d at least have 15-20% market share of the GPU space right now, if not more.
Not only did AMD do all that in the past, but nowadays they refuse to take back market share with better pricing. Their CPU division is absolutely raking it in on earnings sheets, and they are happy to sell GPUs at NVIDIA minus a bit.
I genuinely blame AMD even more than anyone else for the way the GPU market is. If you have been following along as long as I have and seen both companies behavior I don’t see how you could come to a different conclusion. I LOVE talking hardware and am legitimately down for someone to change my mind.
I know AMD is a favorite here on Reddit, and this opinion may make someone mad, but if at least one person reads this and changes their mind it was worth typing. AMD almost killed BOTH CPUs and GPUs. If it wasn’t for Intel’s confusing stagnation, Ryzen wouldn’t have pressed CPUs into a great era of competition. Intel would have the exact same position as NVIDIA, and BOTH situations in that scenario would be caused by AMDs failings. They won the CPU battle for now mostly out of Intel’s failings but I am glad they caught up and started winning there. It’s best for the market if two companies trade blows.
I hope everyone understands that CPUs could be just as bad as GPUs right now if Intel wouldn’t have stagnated. It’s actually good they did imo. AMD’s acquisition of ATI was a huge mistake and the world is a worse place because of it. NVIDIA would not be able to literally be a monopoly and the most valuable company in the world if they had competition, but they currently don’t. There is no competition for their AI cards or their halo products.