This is not investment advice. The author has no position in any of the stocks mentioned. Wccftech.com has a disclosure and ethics policy.

Susquehanna today updated its guidance on AMD (NASDAQ:AMD) stock, upgrading it from sell to a neutral rating and the stock jumped as a result (up about 3% for the day). This came from several factors, but perhaps none more so than concerns around Intel (NASDAQ:INTC) and its continued struggles with 10nm, as Susquehanna cited AMD's improved server class CPU sales.

As many will know, on Intel's earnings call last month, the company revealed that it had delayed volume production on its 10 nanometre manufacturing process until next year while AMD said on its call that it planned to start its next generation 7 nm production in late 2018.

Christopher Rolland published the note to clients today stating "We believe Intel's delay will help to maintain/improve AMD's competitiveness for their next generation of EPYC and Ryzen products", while also later commenting that "For the first time in memory, AMD will compete at a similar process technology as Intel, a strong multi-year tailwind".

It's worth noting that winning business in the server market is so desired as many server providers operate on long sales cycles but also on identical hardware replacement contracts, hence the multi-year comment, a major windfall in addition to the higher pricing associated with this segment.

Analyst U-Turn on AMD

Susquehanna just a couple of short months ago downgraded AMD, at the time citing the Bitmain ASIC for crypto mining as being a major headwind for AMD GPUs. Rolland mentioned in his note that "A lot can change in 2 months" and the upgrade to neutral with a price target that still represents a 12% downside from yesterday's closing price seems... at least somewhat conflicted. Bitmain allegedly "destroyed its value proposition" for the new ASIC by hiking costs since Susquehanna first saw it at $800 compared to the new pricing at $2,150.

Of course, readers in the know will also have concerns about the discussions for a hard fork which seem to semi-regularly swirl around Ethereum to prevent ASICs from being effective, however last month, Vitalik seemingly put the matter at least somewhat to bed stating that a hard fork would be problematic: "Getting everybody to upgrade is likely to be fairly chaotic and detract from more important things. So, at this point I personally lean quite significantly towards no action."

It's important given that additionally, the proposed fork at the time wouldn't make Ethereum forever ASIC resistant, just that it would render existing hardware ineffective, basically sounding the starting gun on an arms race between ASIC makers and Ethereum developers to see who ran out of the will to fight the battle first.

Leading vs. Lagging Indicators...

The analyst game is an intense one at times. People spend their full time jobs looking into all the things they think will have an effect on a company's prospects and stock price. The best ones consistently predict in the right direction and invariably some get it wrong. Ultimately what we're looking at here is a lagging indicator in the Susquehanna analysis. The downgrade on the 26th of March prompted a decline in the stock of another few percent, but only came about when the stock had already fallen by almost 15% in the preceding weeks, while the upgrade now has again prompted a market reaction in the same direction, but comes hot on the heels of a month long rally to bring us back almost full circle.

Keep in mind also that the discussion around 7 nm and 10 nm is not exactly comparing apples to apples, since despite what things may appear to be on the surface, not all nanometres are created equally. Tech savvy readers here will no doubt know that the headline numbers do not make for all the facts, however, at least in this respect, Rolland is somewhat on the money in that the 7 nm in question is likely comparable to the 10 nm of Intel.