• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD on the road to recovery.

NinjaCool said:
AMD sure picked the right partner at the right time as TSMC seems to be soaring ahead :eek:
Click to expand...

I don't think there was any luck involved, to be honest. GloFo used to be AMD's so they'll be fully aware of what they can and cannot do, and GloFo cancelling a process node is more precedent than unusual for them. How good TSMC's 7nm has proven to be has been a pleasant surprise for AMD (as illustrated by Vega 20 and EPYC 2 moving forward), but AMD have known for quite a while that it was always TSMC that would be their 7nm producer.
 
True but one thing that has been lucky for AMD and that is intel becoming stuck in a 10nm development hell!
Remember the 7nm Ryzen was supposed to be fighting against Intel's 10nm Icelake but that sure as heck is not going to happen now :p
 
Last edited:
NinjaCool said:
True but one thing that has been lucky for AMD and that is intel becoming stuck in a 10nm development hell!
Remember the 7nm Ryzen was supposed to be fighting again Intel's 10nm Icelake but that sure as heck is not going to happen now :p
Click to expand...

Was Bulldozer AMD's bad luck? no, it was incompetence and bad decisions.

What you might call AMD luck, others call Intel incompetence.

Intel's inability to compete with TSMC and AMD in technology for that matter is on Intel.
 
According to various articles I've seen, they were likely too ambitious with their jump from 14nm to 10nm - they could've done a more simple 10nm implementation and then gone onto 10nm+ or whatever.

The larger issue for Intel is that they rested on their laurels too much during the Bulldozer era. They had a significant advantage but didn't continue to push ahead. Bad for us then, good for us now.
 
humbug said:
AMD shares are up 45% since i made this thread. its only been up about 6 weeks.
Click to expand...

You didn't believe me when i said it will go $30 by end of the week... Already at $29+ today :)

Jumper118 said:
It's too fast. It will probably crash and settle around low $20's
Click to expand...

Actually the share is on buy by all (around 15) but 4 of the analyst having it on sell. While it was trading at $30 outside the stock market 3 weeks ago!!!!

Yes it will have some correcting drops like this morning, however it seems AMD is going to overcome their own set limit of 5% server market share this year.
Which seems correct, given the amount of virtual server companies facing severe security issues with Intel Xeon which only solution is to shut down Hyperthreading completely crippling their systems performance and capabilities.

On the other side, the company is more than 6 months ahead of it's datacenter competitor, Nvidia, which has not a single product in the pipeline to compete with current AMD products (like V340) nor upcoming products (Vega 20 in single and dual form).

An announcement shortly of the abolition of the penalties has to pay to GF, since the latter cannot make 7nm products, by using TSMC, will add few hundred million USD more to the company coffers annually and all way to 2024, than previously assumed.
 
Jumper118 said:
It's too fast. It will probably crash and settle around low $20's
Click to expand...
Certainly there's going to be some dips.
But with AMD's current product range potential growth in lucrative datacenter/server markets is huge.
And 7nm Zen2 is going to give heck lot more for those markets with its "core density" increase of chips.

And even if/when Intel manages to get their 10nm somewhat working they'll have problems in answering AMD's modular design to scale core counts.
Huge monolithic chips will simply always cost lot.
And Intel can't yet even get small chips working from that node, so it will be long time before it's good for big chips.
That puts AMD into very good position to increase market share in coming years.
 
Jumper118 said:
It's too fast. It will probably crash and settle around low $20's
Click to expand...
May correct a but then consolidate for a while. Earnings will need to catch up with the share price at some point to prove the shares are worth their money. At the moment people are expecting good growth for a while yet. Looking decent for a fairly long term hold IMO (but please dont take this as investment advice - lots of risks of course)
AMD shares have been a bit more speculative in the past so step rises often followed by similar falls if memory serves me correctly. I think things are different under the present leadership however.
 
Panos said:
An announcement shortly of the abolition of the penalties has to pay to GF, since the latter cannot make 7nm products, by using TSMC, will add few hundred million USD more to the company coffers annually and all way to 2024, than previously assumed.
Click to expand...

This is a good spot and will help AMD as up until now they’ve had to pay Global Foundries a fee for every wafer manufactured elsewhere.

AdoredTV did a good video on this: https://youtu.be/kc2GgP5_1J8
 
Jumper118 said:
10% in one day is too fast. 6% the day before. its due a correction.
Click to expand...

That assumes it wasn't horrendously poorly valued for what was coming and that there is no actual reason for the price to increase. The extremely likely massive growth to come in servers in the next 18 months, the growth in desktop the renewed sales from a stronger console market with a likely PS5 in 2019 which would also likely increase Xbox One X sales as with both companies with powerful consoles you'll get more game devs designing more demanding games which may struggle or not be released on the older consoles now.

It's also based on AMD's potential competitiveness with Intel longer term. Intel have lost their process advantage and it's unlikely to come back any time soon. In fact rumours are that Intel are also struggling with EUV as well as their main 10nm process while Samsung are planning the most comprehensive EUV process the soonest with TSMC using a limited EUV as early as late 2019.

Intel are at a fairly critical point, if they can't get their processes back on track they may well have a process node disadvantage for years to come and they simply don't have the volume to invest as heavily in process nodes and other newer technologies as TSMC do. If anything the future as it looks right now is that Intel might have to leave the fab business themselves, which would likely mean selling their fabs to TSMC or Samsung or someone with insanely deep pockets doing the same as Abu Dhabi did and trying to establish a new player. Buy Intel fabs, buy Glofo maybe as well and try to establish a TSMC competitor... not very likely at this point.

Essentially AMD stock was priced as if Intel was completely unbeatable no matter the circumstances and they would only ever have scraps. AMD stock is now being upgraded based on the fact that Intel aren't incapable of huge mistakes, can and do screw up massively and too often and will absolutely lose a fairly hefty market share in multiple markets.
 
Jumper118 said:
10% in one day is too fast. 6% the day before. its due a correction.
Click to expand...

Intel plagued with vulnerabilities in their server chips, microcode and software mitigations crippling their performance, only a hardware redesign can fix the issue, but there is no competitive node process any more and their existing node is already strained with capacity. Meanwhile, AMD has a new CPU architecture that doesn't have anywhere near the same problems and is launching a new server CPU on a cutting edge process at a time when existing server customers are questioning the security and performance of their infrastructure.

And you think AMD's stock price increase is somehow too fast?
 
You must log in or register to reply here.
Back
Top Bottom