# My Quantum Computing Wish List for Christmas 2021 and New Year 2022

Here are my top wishes for developments in quantum computing for Christmas in 2021. Okay, that’s too tall an order with too little time left, so this informal paper lists the quantum computing developments I really want to see in the coming year, 2022. It’s a fun list, but it’s also a real list — these are advances that have a realistic chance of occurring over the coming year. It also gives a sense of how much work remains before quantum computing will really be ready to finally tackle practical real-world problems at production scale.

For reference, here are my Christmas wish lists from the past two years:

*My Quantum Computing Wish List for Christmas 2020 and New Year 2021**My Quantum Computing Wish List for Christmas 2019 and New Year 2020*

I still want everything from my 2019 and 2020 wish lists since so few of my wishes came true, but this year I tried to come up with a more abbreviated list which was a little more practical and achievable.

And may 2022 bring great tidings of comfort and joy!

**Topics discussed in this paper:**

- My top ten Christmas 2021 and New Year 2022 wishes
- A few additional wishes
- A special additional wish
- I know it’s Christmas, but enough with the toy algorithms
- My top five Christmas 2021 wishes?
- No, I’m not wishing for more qubits!
- But… we do need dramatic physical qubit growth to eventually support quantum error correction (QEC)
- I’d rather see better qubits for a smaller quantum computer
- Yes, I want to see more qubits, but not as the highest priority
- Yes, quantum error correction (QEC) is an urgent research priority but near-perfect qubits are a more-urgent near-term priority
- Quantum error correction (QEC) will no longer be high on my wish list until after we get near-perfect qubits
- Of course it will be interesting to see progress with Quantum Volume
- My non-wishes for Christmas 2021 and New Year 2022
- Variational methods are a technical dead-end and unlikely to ever achieve any significant quantum advantage
- My top Christmas 2021 wish
- Quantum Fourier transform (QFT) and quantum phase estimation (QPE) to support quantum parallelism are the Holy Grail for quantum computing
- My alternative top Christmas 2021 wish
- My other alternative top Christmas 2021 wish
- Some successes on my 2019 and 2020 wishes
- Some notable failures on my 2019 and 2020 wishes
- Still a mere laboratory curiosity of interest primarily to The Lunatic Fringe
- Still in pre-commercialization, still years from commercialization
- State of the art for quantum computing
- Changes to my wishes?

# My top ten Christmas 2021 and New Year 2022 wishes

My top ten Christmas 2021 and New Year 2022 wishes, not in any strict order per se:

**Qubit fidelity of 3.5 nines.**Approaching*near-perfect qubits*. I really want to see four nines (99.99%) within two years, but even IBM is saying that they won’t be able to deliver that until 2024. Trapped-ions may do better.**Hardware capable of supporting quantum Fourier transform (QFT) and quantum phase estimation (QPE).**Hopefully for at least 20 to 30 qubits, but even 16 or 12 or 10 or even 8 qubits would be better than nothing. Support for 40 to 50 if not 64 qubits would be my more ideal wish, but just feels way too impractical at this stage. QFT and QPE are the only viable path to quantum parallelism and ultimately to dramatic quantum advantage.**Fine granularity of phase.**Any improvement or even an attempt to characterize phase granularity at all. My real wish is for at least a million to a billion gradations — to support 20 to 30 qubits in a single QFT or QPE, but I’m not holding my breath for this in 2022.**At least a handful of automatically scalable 40-qubit algorithms.**And plenty of 32-qubit algorithms. Focused on simulation rather than real hardware since qubit quality is too low. Hopefully using quantum Fourier transform (QFT) and quantum phase estimation (QPE).**Simulation of 44 qubits.**I really want to see 48-qubit simulation — and plans for research for 50 and 54-qubit simulation. Including support for deeper circuits, not just raw qubit count. And configurability to closely match real and expected quantum hardware over the next few to five to seven years.**At least a few programming model improvements.**Or at least some research projects initiated.**At least a handful of robust algorithmic building blocks.**Which are applicable to the majority of quantum algorithms.**At least one new qubit technology.**At least one a year until we find one that really does the job.**At least some notable progress on research for quantum error correction (QEC) and logical qubits.**Possibly a roadmap for logical qubit capacity. When can we see even a single logical qubit and then two logical qubits, and then five logical qubits and then eight logical qubits?**A strong uptick in research spending.**More new programs and projects as well as more spending for existing programs and projects.

# A few additional wishes

Some items that I really want but didn’t fit in the count:

**At least a paper proposal for an application to reach****The ENIAC Moment****.****Greater transparency.**What’s really going on under the hood of these quantum computers?! And why can’t the trapped-ion vendors be much more transparent?!**Better documentation and specifications.****Better roadmaps.**More technical details and milestone metrics. See my critique of IBM’s quantum roadmap.**Some fruit from Amazon’s research.**More than just AWS hosting services for quantum computers from other vendors. An actual quantum computer.**Trapped-ion vendors catch up with transmon vendors.**I’d like to see 27 and 32-qubit trapped-ion quantum computers. And a clear and detailed roadmap for 40, 48, 50, 64, 72, 80, 96, and 128-qubit trapped-ion quantum computers. And transparency on qubit fidelity, fine granularity of phase, circuit depths, and gate execution times.**Analytic tools to detect scalability problems for algorithms.**Be able to detect coding patterns in algorithms which won’t scale to run on particular hardware configurations expected over the next few to five to seven years. Scaling for both up to 48 to 50 qubits and in the over-50 qubit range as well. Including circuit depth and coherence time. And this includes fine granularity of phase and probability amplitude to detect overly-ambitious use of quantum Fourier transform (QFT) and quantum phase estimation (QPE), and quantum amplitude estimation (QSE) as well.

# A special additional wish

More a wish for me personally:

**I’d like to learn more about how quantum logic gates are implemented at the firmware and hardware level.**Especially the actual physics. I’d like to understand theoretically how fine-grained rotations of phase angles and probability amplitudes can really be — how much is a theoretical limit vs. how much is just a limit of the current firmware, digital logic, and analog control logic. My motivation is to understand the implementation and limitations of quantum Fourier transform (QFT) and quantum phase estimation (QPE) in support of quantum computational chemistry and other applications for QFT and QPE.

# I know it’s Christmas, but enough with the toy algorithms

Very small algorithms can be illustrious for beginners, but the field is *plagued* by tiny algorithms. I don’t care what you can do with five or seven or even eleven qubits. If your algorithm isn’t using at least 24 qubits, then I don’t want to hear about it. Okay, maybe that’s a bit extreme, but at least 20 qubits or maybe 16 qubits at a minimum. I’m at least comfortable saying that anything less than a 16-qubit quantum algorithm is definitely a *toy algorithm*.

Granted, current hardware may not support algorithms over six, seven, eight, or ten qubits (Quantum Volume of 64, 128, 256, or 1024), but I’d rather see a simulation of a significant algorithm than an actual run of a mediocre algorithm on current hardware.

We need to be preparing for the hardware which will be available in two to three years, not *proving* that current hardware is mediocre.

To be clear, I’m promoting *scalable* quantum algorithms — *automatically* scalable quantum algorithms. It’s okay if you want to demonstrate that your 24 or 32-qubit algorithm can also scale *down* to 20, 16, 12, or even 8 qubits to run on current hardware, but it is *scaling up* to 32, 36, and 40 qubits which really matters most.

# My top five Christmas 2021 wishes?

I really tried to whittle my wish list down from ten to five, but I really can’t. You can pick any five of my top ten at random, but I can’t do any better than that.

# No, I’m not wishing for more qubits!

Sure, a 256-qubit quantum computer, or 160 qubits, would be great, but the simple fact is that we have reached the point where raw qubit count is not the limiting factor. The main limiting factors are:

**Qubit fidelity.**Really need four nines to do anything substantial.**Coherence time and circuit depth.**Too short to do much of value.**Gate fidelity.**Gate error rate is still too high.**Measurement fidelity.**Qubit results mean nothing if you can’t accurately measure them.**Connectivity.**Full any to any connectivity is needed. Or at least something much closer than nearest neighbor.**Automatically scalable 40-qubit algorithms.**See we need even larger and more complex algorithms, but we can’t even muster 40 qubits at this time.

So, unless there is dramatic (or even moderate) progress on all of those technical criteria, any increase in qubits (we already have 53, 65, and now 127 qubits) will be moot.

Seriously, although both Google and IBM achieved 53 qubits and IBM achieved 65 qubits, there is *no evidence* that the availability of this hardware has advanced algorithm design, although Google themselves did post several papers, but they used no more than 23 qubits.

Until we achieve major advances on *all* of those fronts, even 48 qubits are more than we need at this time.

# But… we do need dramatic physical qubit growth to eventually support quantum error correction (QEC)

Although additional noisy qubits aren’t helpful to algorithm designers and application developers, we do need to eventually get to much larger numbers of physical qubits, even if still somewhat noisy, to support quantum error correction (QEC) and logical qubits.

For example, if each logical qubit required 65 physical qubits (one IBM approach), even 48 logical qubits would require 48 x 65 or 3,120 physical qubits. With only 1,121 physical qubits, IBM Condor would support only 1,121 / 65 or 17 logical qubits, which wouldn’t be terribly useful. To support 40-qubit algorithms, 40 x 65 or 2,600 physical qubits would be needed.

So, I do indeed wish to see dramatic qubit growth over the coming year, but not as high a priority as pushing towards the *near-perfect qubits* which will be more useful over the next few years.

This is an example of research being at different time scales — near term, medium term, and longer term. We need near-perfect qubits in the near and medium term but logical qubits in the medium to longer term.

# I’d rather see better qubits for a smaller quantum computer

Rather than larger qubit counts, such as more 100-qubit machines and even a 256-qubit machine, I’d rather see a 48-qubit machine with better qubits. And even refinements for 28 and 32-qubit machines.

# Yes, I want to see more qubits, but not as the highest priority

Eventually we will indeed need 256 and 1,000-qubit machines, and even 4K-qubit machines, but until qubit quality permits larger and more complex quantum algorithms, those higher qubit counts will not be used effectively.

# Yes, quantum error correction (QEC) is an urgent research priority but near-perfect qubits are a more-urgent near-term priority

Quantum error correction (QEC) is still years away. Yes, research is needed and I expect progress in research over the coming year, but I expect nothing to put in the hands of users for another couple of years. Quantum algorithms need significant near-term improvements in qubit quality, so I would give near-perfect qubits a higher near-term priority.

Besides, higher-quality physical qubits make it easier to achieve quantum error correction — with significantly fewer physical qubits per logical qubit.

# Quantum error correction (QEC) will no longer be high on my wish list until after we get near-perfect qubits

Progress on quantum error correction (QEC) and logical qubits, also known as *fault-tolerant quantum computing*, was previously high on my Christmas wish lists, and yes, quantum error correction (QEC) must remain a high research priority, but as a longer-term goal, without usable results in the next couple of years. It won’t be high on my wish list until after we have near-perfect qubits available, such as four nines of 40 qubits.

This is consistent with my interest in focusing quantum algorithms on simulation rather than real hardware for the next few years, and even then, focusing on near-perfect qubits until QEC becomes a mainstream feature of quantum computers.

# Of course it will be interesting to see progress with Quantum Volume

It’s kind of a slam dunk for IBM to achieve a Quantum Volume (QV) of 256 in 2022 — they’re at 128 now. I would hope they could easily hit 512 (9 qubits). Maybe even 1K (10 qubits). But 2K or 4K or higher are probably out of reach this coming year, or maybe even the following year.

I do worry a bit about IBM since they have only very weak connectivity. Their achievement of QV of 128 — basically 7 qubits — relies on their heavy-hexagon topology. Moving to higher QV will incur long SWAP networks, which will require significantly greater qubit fidelity.

What Quantum Volume is reached by the trapped-ion vendors is another matter. But, again, Quantum Volume isn’t a priority on my wish list. It’s just a curiosity.

And, to be clear, whatever Quantum Volume that gets reported should be *measured*, not merely *estimated*.

IonQ announced a 32-qubit trapped-ion quantum computer with 22 *algorithmic qubits* (AQ) which they *estimated* would have a Quantum Volume “*of at least 4,000,000*.” Benchmarks must be *measured*, not *estimated*.

Honeywell *measured* QV of 1024 with 10 trapped-ion qubits. I expect they will improve on that in 2022.

# My non-wishes for Christmas 2021 and New Year 2022

Just to recap, here are perfectly reasonable achievements which quantum computing could make in the coming year, but they simply aren’t in my personal top ten priorities at this time:

**More qubits.****Progress in Quantum Volume (QV).**Not a particularly helpful metric at this stage. The focus should be on enhancing qubit quality and qubit-specific metrics such as nines of qubit fidelity and connectivity.**Progress on quantum error correction (QEC) and logical qubits.**Research, yes, but nothing to put in the hands of users for a few more years.**More hardware vendors.**Nominally I do still want to see more hardware vendors, but I want to see more hardware progress from the existing hardware vendors first as the higher priority.

# Variational methods are a technical dead-end and unlikely to ever achieve any significant quantum advantage

Although variational methods are quite popular, they are unlikely to ever achieve any significant quantum advantage. Without a significant quantum advantage, they are a technical dead-end.

They only succeed by breaking a problem down into much smaller pieces, but that also reduces any advantage by reducing the extent of any quantum parallelism.

My point here is that I would like to see advances in quantum algorithms and quantum applications, but use of variational methods will undermine if not absolutely eliminate the advantages of such quantum algorithms and quantum applications.

Quantum algorithm design should focus primarily on quantum Fourier transform (QFT) and quantum phase estimation (QPE), not variational methods. Sure, this means running on simulators rather than real quantum computers, but this is the approach which will eventually achieve *dramatic quantum advantage*, not variational methods.

So my wish for Christmas and the New Year is for algorithm designers to move away from use of variational methods on current limited hardware in favor of quantum Fourier transform (QFT) and quantum phase estimation (QPE) on simulators.

# My top Christmas 2021 wish

If I could have only one wish, what would it be? I’d wish for…

**Hardware capable of supporting quantum Fourier transform (QFT) and quantum phase estimation (QPE).**Hopefully for at least 20 to 30 qubits, but even 16 or 12 or 10 or even 8 qubits would be better than nothing. Support for 40 to 50 if not 64 qubits would be my more ideal wish, but just feels way too impractical at this stage. QFT and QPE are the only viable path to quantum parallelism and ultimately to dramatic quantum advantage.

This wish would greatly facilitate support for quantum computational chemistry and allow us to move away from the unproductive distraction of variational methods, which would never achieve dramatic quantum advantage anyway.

This is actually a clever wish because achieving it requires dramatic advances on a number of fronts:

**Qubit fidelity.**Low error rate.**Qubit connectivity.**Much better than only nearest neighbor. Sorry, SWAP networks don’t cut it.**Gate fidelity.**Low error rate. Especially 2-qubit gates.**Measurement fidelity.**Low error rate.**Coherence time and circuit depth.**Can’t do much with current hardware.**Fine granularity of phase.**At least a million to a billion gradations — to support 20 to 30 qubits in a single QFT or QPE.

# Quantum Fourier transform (QFT) and quantum phase estimation (QPE) to support quantum parallelism are the Holy Grail for quantum computing

So far, quantum Fourier transform (QFT) and quantum phase estimation (QPE) are the best known technical approach to achieving quantum parallelism. That makes them the Holy Grail of quantum computing. QFT and QPE are the only viable path to quantum parallelism and ultimately to dramatic quantum advantage.

Maybe some other approaches will be surfaced by research in the coming years, but for now they are the best we have. But they are not getting enough attention.

# My alternative top Christmas 2021 wish

Since my top wish may seem more like a cheat since I’m supposed to be asking for only one wish, I’d replace that wish with:

**Fine granularity of phase.**At least a million to a billion gradations — to support 20 to 30 qubits in a single QFT or QPE.

That’s still a fairly clever wish since achieving it might result in many of the other qualities required for QFT and QPE being achieved as a side effect of general improvement of qubit fidelity.

# My other alternative top Christmas 2021 wish

And if even that seems like too much of a cheat, I’d wish for simply:

**Qubit fidelity of 3.5 nines.**Approaching*near-perfect qubits*. I really want to see four nines (99.99%) within two years, but even IBM is saying that they won’t be able to deliver that until 2024. Trapped-ions may do better.

# Some successes on my 2019 and 2020 wishes

Although most of my wishes from prior years have not been granted, there have been a few successes:

**100+ qubits.**At least one machine in the 100+ qubit range. IBM just announced the 127-qubit Eagle. Another 100+ machine, Atom Computing Phoenix, has been “unveiled”, but its actual availability is unclear. And another, ColdQuanta Hilbert, is expected early in 2022.**Additional hardware vendors.**More than a few have popped up.**Another qubit technology.**Two vendors of neutral atom qubits have surfaced.**Much more fundamental R&D.**A palpable increase and announcement of new programs and projects. Not as much as I wanted, but some progress.**Honeywell has made notable progress with QV of 1024.**That at least demonstrates that trapped-ion qubits have superior connectivity compared to superconducting transmon qubits.

# Some notable failures on my 2019 and 2020 wishes

**No quantum computer from Microsoft.****No quantum computer from Intel.****No new quantum computer from Google.****Rigetti still hasn’t gotten past 32 qubits.****IonQ hasn’t gotten past 32 qubits.****Xanadu is still far from a practical photonic quantum computer.**Great promise, but we’re trying to move past mere promises.**Haven’t gotten to three nines of qubit fidelity.**IBM did announce something recently, but it was “best” case, not nominal case.**No improvement in qubit connectivity.**For superconducting transmon qubits. Trapped-ion qubits seem all set.**Simulators haven’t gotten past 32 or 40 qubits.****People are still touting Grover search and Shor’s factoring algorithm.**Time to move on! Focus on practical algorithms for the three to five year horizon.**Limited transparency.****Mediocre documentation.****Mediocre roadmaps.****No apparent progress on fine granularity of phase or quantum Fourier transform (QFT) or quantum phase estimation (QPE).**

# Still a mere laboratory curiosity of interest primarily to The Lunatic Fringe

I know people are working really hard and great progress is being made, but even with *all* of my Christmas and New Year wishes, quantum computing will still remain *a mere laboratory curiosity* of interest primarily to The Lunatic Fringe — who will work with any technology no matter how limited or how far from completion it is — even as 2022 ends and 2023 begins.

We probably need another two or three years. Or four or five. Or more.

And we probably need full, automated, transparent quantum error correction (QEC) — or near-perfect qubits (four to five nines) which are close enough to perfect that QEC is not needed for many algorithms and applications.

# Still in pre-commercialization, still years from commercialization

Much research, prototyping, and experimentation is still required until we have sufficient technical knowledge and capabilities to even begin commercialization of quantum computing, which itself will take years.

All of my Christmas and New Year wishes are made in the spirit of pre-commercialization — none are intended to imply immediate or even eventual availability in commercial products.

For more on research needed, see my paper:

*Essential and Urgent Research Areas for Quantum Computing*- https://jackkrupansky.medium.com/essential-and-urgent-research-areas-for-quantum-computing-302172b12176

For more on pre-commercialization, see my paper:

*Model for Pre-commercialization Required Before Quantum Computing Is Ready for Commercialization*- https://jackkrupansky.medium.com/model-for-pre-commercialization-required-before-quantum-computing-is-ready-for-commercialization-689651c7398a

Or for a summary of both:

*Prescription for Advancing Quantum Computing Much More Rapidly: Hold Off on Commercialization but Double Down on Pre-commercialization*- https://jackkrupansky.medium.com/prescription-for-advancing-quantum-computing-much-more-rapidly-hold-off-on-commercialization-but-28d1128166a

# State of the art for quantum computing

My list of wishes can also be read as a summary of the *state of the art* for quantum computing — not so much how much has been done, but how much work remains until quantum computing will be ready for development and deployment of production-scale practical real-world applications which can routinely achieve dramatic quantum advantage.

Hint: *Much* work remains.

# Changes to my wishes?

I reserve the right to make changes to my wishes right up until 11:59 PM on December 24, 2021 — Christmas Eve. And I expect Santa’s elves to be responsive to such late-breaking requests!

For more of my writing: ** List of My Papers on Quantum Computing**.