Sat, Oct 25, 2014, 3:24 PM EDT - U.S. Markets closed


% | $
Quotes you view appear here for quick access.

NVIDIA Corporation Message Board

schmenkle321 7 posts  |  Last Activity: Oct 9, 2014 9:13 AM Member since: Jul 13, 2007
SortNewest  |  Oldest  |  Highest Rated Expand all messages
  • schmenkle321 schmenkle321 Oct 9, 2014 9:13 AM Flag

    You've said NVDA will be a much bigger company, so I assume you have a long term term view - NVDA has had more impact on AMD than vise versa. Note the purchase of ATI. I've said for a long time AMD can't fight wars on two fronts with two formidable companies. ATI was purchased for $5.4B, today their market cap is well under 1/2 that. Huge mistake. But AMD saw the value of a graphics business and had to have one. As a result they are carrying a huge debt burden and that will ultimately drag them down. Earnings are going to suck so they may not be able to service that debt. That could trigger bankruptcy. Read's departure was hasty with little explanation, not a good sign. That means near term piles of doo doo for AMD to navigate through. Earnings are scheduled for next week. I'd guess revenue miss, non optimistic guidance and a whole lot of words around restructuring and new vision.

    What does that mean for NVDA? It looks to me like a positive, both short and long term. Customers (OEMs, ODMs, consumers) will have new questions near term about purchasing AMD products and longer term support/roadmap. That should push more business NVDA's way. But I also think NVDA is unfairly compared to AMD as a peer -- which makes sense only with each company's GPU business. I feel AMD's market outlook against Intel is reflected on to NVDA's outlook. But NVDA and Intel really are not competitors in their core businesses, more like cooperative antagonists.

    Short term, I really don't care what the market does. My cynical expectation is the market will open with the usual knee jerk selloff around any uncertainty with NVDA. It will take a few minutes/hours/days to figure out this is really a positive. But also cynically, I don't expect any positive reflection in stock price either. That will only come with continued long term performance. Long term, longs should thank AMD for shooting themselves in the foot for us.

  • schmenkle321 schmenkle321 Sep 11, 2014 12:58 AM Flag

    I'll bite, some random thoughts:

    - hurrah to apple for being the first to debug TSMC's 20nm process.

    - The fact the iPh6 ship date is "in the future" could be troubling for AAPL shareholders, they have effectively stopped sales of iph5 with yesterday's announcement, so any TSMC production hiccoughs will be reflected in the share price.

    - size, process, and cost in comparison to other parts are really meaningless as apple is a captive user.

    - Take the improvement numbers mentioned with a grain of salt. There will certainly be improvements and certain pipelines *may* have been improved to the factor reported. But these are not linear relationships to benchmarks or application performance. Wait for some real world testing before passing judgement.

    - I think the question on the table is how is this part going to stack up against K164. The answer is it's too early to know. Core arch is a push. Process is a push. Design teams donno. First to market apple. Predictive algorithm may give a nod to NVDA in CPU perf. GPU perf should go to NVDA. Power consumption donno but my guess is apple is conservative here so may have an edge, but I'm not an engineer. As I say, need to get the part in reputable hands and test it.

    It will be interesting to see how AAPLs part stacks up against K132 as a proofpoint. Process alone aapl will have a lead, but if it doesn't, particularly in GPU, it could be a bummer for apple. AAPL has never been about speeds and feeds, so I could see a scenario with them losing every major benchmark and saying meh, they will still sell a lot of phones (though I'm not too certain about how many watches).

  • schmenkle321 schmenkle321 Sep 5, 2014 10:30 AM Flag

    CUDA is a programming language that allows the parallel processing that is inherent in the GPU to be purposed for computations other than graphics. "General Purpose" GPU is the idea. Before Cuda a standard general large parallel processing programming language didn't exist. If you really want to learn more about it, the wikipedia entry is a good overview. GPUs can perform certain functions the CPU used to perform a whole lot more efficiently.

    As for implications, the concept of parallel processing can be challenging as traditionally programmers thought about problem solving in a more serial way. So NVIDIA embarked on a campaign to publish and teach the concept in the mid 2000s. They lectured and worked with major universities to begin to teach CS students the idea of opening up these computational resources (and to sell more of their products). If you want some great examples of GPGPU, look at: nvidiaDOTcom/object/tesla-case-studies.html

    The long term implications can be big, but CUDA won't propel NVIDIA to greatness alone. GPGPU has enhanced certain tasks (climate modeling, medical imaging, Black Sholes modeling) but it is overkill for most problems. The setup penalty is huge for a small number of calculation for example, and a CPU could do it in a fraction of the cycles. Where GPGPU really shines is on large data sets, and not every problem is a large data set.

    The potential as I see it is that a traditional CPU becomes a smaller silicon budget in a system while GPUs take more real estate. Not only can GPUs do graphics, they do complex calculations too. And that is goodness for NVDA. But getting the rest of the world on that page is not simple. But I have no doubt NVIDIA will continue to push this technology and I am very interested to see if and when consumer based GPGPU applications become a reality.

  • schmenkle321 by schmenkle321 Aug 8, 2014 2:55 AM Flag

    A coherent discussion with you is not possible. Why don't you stop trying to pick a fight and I won't reply and we'll be good. cheers

  • Reply to

    I still say we'll see 19+

    by kvimax1 Aug 7, 2014 6:34 PM
    schmenkle321 schmenkle321 Aug 8, 2014 1:01 AM Flag

    The topic was NVDA's licensing business, not Intel's CPU IP. I have no idea what patent's of Intel's NVDA is potentially stepping on. If you're saying Intel holds big cards here, I say ARM seems to be doing pretty well without an Intel cross license.

    Why don't you can enlighten us about what ARMv8-A IP intel has claim to?

  • Reply to

    I still say we'll see 19+

    by kvimax1 Aug 7, 2014 6:34 PM
    schmenkle321 schmenkle321 Aug 7, 2014 10:28 PM Flag

    Hey Max, always appreciate your comments and insight. I agree new highs are headed our way, there is just too much right going on here.

    One comment on the BofA guy. I went back and listened and yes I agree he is trying to interject some FUD. But his question is not exposing a weakness though he may have desired or intended it to. Licensing is a very, very, very good thing IMO and Intel HAS TO re-up. INTC are building NVDA IP into every integrated graphics part they build, and they set a precedent they (and the rest of the market for that matter) MUST to pay for it if they use it. Intel has guaranteed this straight-to-the-bottom-line profit for NVDA for years to come. And NVDA basically doesn't need to lift a finger to bank it. Hey, a lot can happen in the next 2 years, but Intel needing graphics (and infringing on NVDA IP), that is about a sure a thing as there is.

    re: "They have literally INVENTED new markets for themselves and NO ONE really understands it, YET."

    Only guys who've been around for 15 years (and maybe a few others) have the proper perspective. :)

  • Reply to

    I still say we'll see 19+

    by kvimax1 Aug 7, 2014 6:34 PM
    schmenkle321 schmenkle321 Aug 7, 2014 6:52 PM Flag

    always appreciate your optimism Max. Multiple posts including yours about NVDA being "hated" so I'm not sure the games are over and this stock can get back into the "favorable" column with investors just yet.

    I love the growth story, anyone can see it, NVDA has been undervalued for a long time now. cheers

18.48+0.19(+1.07%)Oct 24 4:00 PMEDT

Trending Tickers

Trending Tickers features significant U.S. stocks showing the most dramatic increase in user interest in Yahoo Finance in the previous hour over historic norms. The list is limited to those equities which trade at least 100,000 shares on an average day and have a market cap of more than $300 million.