REVIEW 3 major objections 4 minor 16 references
Designing a new spatial light modulator for holographic photostimulation
T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read The MacroSLM, a 1536 x 1536 liquid-crystal-on-silicon modulator built for photostimulation, switches 8-spot holograms at 500 Hz at 1064 nm while retaining 96% of its slow-speed diffraction efficiency.
desk verdict A credible engineering demo of a 1536x1536 NIR SLM hitting 500 Hz hologram switching; main weakness is missing temperature-stability data, but it deserves serious review. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing mechanism is overdrive combined with phase wrapping, executed through a custom FPGA data pipeline. Overdrive applies a transient voltage beyond the target so each pixel's phase reaches its destination faster, and phase wrapping lets each pixel choose the fastest among phase destinations that differ by one wave. Because transient frames must be calculated for every pixel on the fly to allow arbitrary switching order, the drive electronics split the 1536 by 1536 array among eight secondary FPGAs that compute overdrive voltages and load them at about 1250 Hz, with a primary FPGA coordinating data distribution and trigger handling. Supporting design choices include a 20 µm pixel pitch that raises fill factor and reduces fringing-field crosstalk, a 0 to 12 V analog drive that enables strong overdrive, and Peltier backplane heating/cooling that keeps the liquid crystal at 45°C.
What would settle it
Set up the same hologram-to-hologram test at 1064 nm while inserting a fast thermocouple or resistance sensor in the liquid-crystal cell and varying the incident NIR power from low to several watts; if the cell temperature drifts away from 45°C or the 500 Hz peak-to-peak amplitude dips below 90% of the 100 Hz baseline, the 500 Hz claim fails.
Extended reading notes
Core claim
On the authors' own terms, the discovery is a full-pipeline demonstration: trigger reception, image calculation, pixel addressing, and liquid-crystal response together deliver a new complex hologram every 2 milliseconds at 1064 nanometers. Speed is defined as the triggered rate at which the system can switch between 8-spot Gerchberg-Saxton holograms while keeping more than 90% of its slow-speed diffraction efficiency; the measured value is 500 Hz with 96% amplitude retention, and 600 Hz with 88% retention. The same device refreshes pixel voltages at 1250 Hz and accepts triggers at arbitrary intervals with 6 µs ±3 µs latency because the image download is interruptible. The authors further argue that visible-wavelength operation should be more than three times faster, since liquid-crystal response slows markedly in the near infrared.
Load-bearing premise
The central claim assumes the backplane Peltier system keeps the liquid crystal at a constant 45°C during high-power laser illumination, because the overdrive timing calibration is strongly temperature-dependent, and the paper reports no direct temperature-stability measurements.
Editorial extensions
If this is right
- At 1064 nm, complex 8-spot holograms can be updated at 500 Hz with more than 90% of slow-speed diffraction efficiency, making closed-loop optogenetics experiments feasible at millisecond timescales.
- Because triggering is interruptible, hologram transitions can be initiated at arbitrary intervals with 6 µs ±3 µs latency, so stimulus timing control no longer depends on integer multiples of a refresh rate.
- At visible wavelengths the liquid-crystal response is expected to be more than three times faster, which would put the same architecture at multi-kilohertz switching speeds.
- The 30.7 mm square aperture and 20 µm pixels reduce the magnification change and optical aberrations when matched to large-back-aperture microscope objectives, and spread high-power illumination over a larger area.
- Multiplexing two such systems has already produced frame rates above 1 kHz, showing the single-device speed is not the ceiling for experiments that can use multiple modulators.
Reading between the lines
- Because the published speed test uses deliberately complex 8-spot Gerchberg-Saxton holograms, simpler holograms or lower spatial frequencies would likely switch faster than 500 Hz, meaning the quoted rate is a conservative operating point for many applications.
- The strong temperature sensitivity of overdrive timing implies the 500 Hz figure transfers to other setups only if the liquid-crystal temperature is held as tightly as asserted; a direct measurement of cell temperature versus incident laser power would reveal the margin.
- The same combination of high-voltage overdrive, interruptible downloads, and large pixels could be applied to visible light or other near-infrared bands for laser micromachining or optical trapping, where switching speed limits throughput.
- A quantitative trade-off study linking pixel pitch, fill factor, and crosstalk to the measured diffraction efficiency would let other groups choose design parameters without building a full modulator.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports the design, construction, and characterization of a 1536×1536 liquid-crystal-on-silicon spatial light modulator (the MacroSLM) with 20 µm pixels, 0–12 V analog addressing, FPGA-based per-pixel overdrive, and Peltier backplane temperature control at 45°C. The central result is a measured hologram-to-hologram frame rate of 500 Hz at 1064 nm, defined as the triggered rate at which 8-spot Gerchberg-Saxton holograms maintain more than 90% of their slow-speed diffraction efficiency; at 500 Hz the measured peak-to-peak amplitude is 96% of the 100 Hz reference, and at 600 Hz it is 88%. The paper also describes the pixel-count and pixel-pitch design tradeoffs, the FPGA data pipeline, and an interruptible download scheme with claimed trigger latency of 6 µs ± 3 µs.
Significance. If the central speed claim holds, this is a substantial advance for holographic photostimulation, which has typically been limited to ~60 Hz commercial SLMs. The operational frame-rate definition based on realistic multi-spot GS holograms at 1064 nm is a sound and appropriately conservative metric, and the effort to measure at the intended operating wavelength rather than at visible wavelengths is commendable. The design choices—large pixels for fill factor and crosstalk reduction, high-voltage drive, phase wrapping, and FPGA-computed overdrive—are well motivated and constitute a useful engineering contribution. The paper explicitly identifies the temperature sensitivity of overdrive timing as a critical calibration issue, which is honest but also highlights the main gap in evidence.
major comments (3)
- [2, 'Backplane heating'] The paper states that 'the timing of our transient overdrive voltages is very temperature-dependent' and that the backplane Peltier maintains the LC at 45°C, but no temperature stability data are provided for either the calibration conditions or the speed test in Section 3. Because the 500 Hz/96% result depends on correctly timed overdrive frames, the absence of any measured temperature record (e.g., t, ΔT during the test, or Peltier setpoint tolerance) leaves the reproducibility of the headline claim unverifiable under varying laser power and self-heating. Please provide temperature measurements during the speed characterization and state the temperature tolerance required for the overdrive calibration to hold.
- [3, Fig. 4] The speed characterization appears to rely on a single photodetector trace at each trigger frequency; no error bars, number of repetitions, or noise floor are reported. Since the stated criterion is >90% of steady-state efficiency and the 500 Hz measurement is 96%, the margin is small enough that measurement uncertainty could alter the conclusion. Please report repeated trials, the variance of the amplitude ratio, and the detector noise, and clarify whether the quoted 96% refers to peak-to-peak photodetector amplitude or to diffraction efficiency (power), since the abstract and Section 4 use the latter while Section 3 uses the former.
- [2, 'Flexible triggering'] The trigger latency of 6 µs ± 3 µs is presented as a capability, but no measurement method or data are shown. Please either provide a description of how latency and jitter were measured (e.g., photodetector versus trigger signal on an oscilloscope) or clearly label the value as a design specification.
minor comments (4)
- [2, 'Voltage range'] The expression for the phase response contains a typographical error: the formula should be φ(t) = φ_i + (φ_f − φ_i)(1 − exp(−(t − t_i)/τ)), but the text shows a garbled denominator. Please correct the equation.
- [3] For consistency with the defined >90% efficiency criterion, please restate the 96% and 88% values as efficiency percentages or explicitly note that the photodetector signal is proportional to optical power, so that amplitude and efficiency are interchangeable.
- [4] The phrase 'triggered at arbitrary intervals of < 1 ms' is ambiguous; please rephrase to clearly indicate that intervals shorter than 1 ms (equivalently rates above 1 kHz) are possible.
- [References] Consider adding the original Gerchberg and Saxton reference (Optik, 1972) alongside reference [8] for completeness of the algorithmic attribution.
Circularity Check
No circularity: the 500 Hz speed claim is a direct measurement against a consistent operational threshold, not a derived prediction.
full rationale
The paper is an engineering report whose central quantitative claim—500 Hz hologram-to-hologram switching at 1064 nm while maintaining 96% of slow-speed diffraction efficiency—is a direct measurement, not a prediction derived from fitted inputs. The frame-rate definition (>90% of steady-state efficiency at a triggered rate) is an operational threshold applied consistently to measured peak-to-peak amplitudes in Figure 4, and the slow-speed 100 Hz reference provides an independent baseline. No parameter is fitted to a subset of data and then renamed as a prediction: the overdrive timing and voltage-phase calibrations are engineering prerequisites, while the reported speed is measured after those calibrations using representative Gerchberg-Saxton holograms. References to prior work by the same group are not load-bearing for the central claim; the only notable one (Ref. 17, MultiSLM) is cited as an application of the system, not as evidence for the speed measurement itself. The paper's own statement that transient overdrive timing is 'very temperature-dependent' identifies a reproducibility gap—no quantitative temperature-stability data are reported—but this does not make the derivation circular, because the headline result is an empirical measurement rather than a quantity that reduces by construction to the calibration. No self-definition, fitted-input-as-prediction, or self-citation chain that forces the result was found, so no circular step meets the evidentiary standard.
Assumptions & free parameters
assumptions (4)
- standard math The diffraction efficiency of a blazed phase grating is η = |sinc(1/Z)|^2, where Z is the number of pixels per grating period.
- domain assumption Liquid crystal phase response to a voltage step is approximately exponential with a time constant depending on transition direction and LC viscosity.
- domain assumption Backplane Peltier heating/cooling maintains the LC at a constant 45°C, keeping viscosity and phase-voltage calibration stable.
- domain assumption The 1064 nm wavelength and the specific 8-spot GS hologram set are representative of photostimulation conditions.
Cite this review
Pith. "Pith review of Designing a new spatial light modulator for holographic photostimulation." pith.science (2026). https://pith.science/paper/ETNZO5PA
@misc{pith2026190804859,
author = {Pith},
title = {Pith review of: Designing a new spatial light modulator for holographic photostimulation},
year = {2026},
howpublished = {\url{https://pith.science/paper/ETNZO5PA}},
note = {Machine review of arXiv:1908.04859}
}
read the original abstract
Driven by the demands for speed and field of view in the holographic photostimulation community, we designed, built, and tested a liquid crystal on silicon (LCoS) spatial light modulator (SLM) with a 1536x1536 square pixel array and high-voltage LC drive. We discuss some of the engineering work that made the MacroSLM possible, including the custom FPGA board for handling huge data rates, the large pixel size for minimizing rolloff and crosstalk, and the temperature control to handle heating effects from the high-voltage controls and high-power laser illumination. We also designed an FPGA implementation of the overdrive method for increasing liquid crystal switching speed, allowing us to overcome the significant data bottlenecks that limit frame rates for large arrays. We demonstrate 500 Hz hologram-to-hologram speed at 1064 nm operating wavelength, and discuss the new science that these speeds and array sizes have enabled.
Reference graph
Works this paper leans on
-
[1]
Two-photon optogenetic toolbox for fast inhibition, excitation and bistable modulation,
Prakash, R., Yizhar, O., Grewe, B., Ramakrishnan, C., Wang, N., Goshen, I., Packer, A. M., Peterka, D. S., Yuste, R., Schnitzer, M. J. and Deisseroth, K., “Two-photon optogenetic toolbox for fast inhibition, excitation and bistable modulation,” Nat. Methods 9(12), 1171–1179 (2012)
work page 2012
-
[2]
data handling. As the pixel count increases, the cost of fabricating the large silicon backplane increases, and yield-related losses can potentially increase. Larger arrays may exceed the single-shot write area of the fabrication foundry, so care must be taken to avoid misalignments and other nonuniformities when stitching together write areas. Data handl...
-
[3]
Simultaneous cellular-resolution optical perturbation and imaging of place cell firing fields,
Rickgauer, J. P., Deisseroth, K. and Tank, D. W., “Simultaneous cellular-resolution optical perturbation and imaging of place cell firing fields,” Nat. Neurosci. 17(12), 1816–1824 (2014)
work page 2014
-
[4]
Packer, A. M., Russell, L. E., Dalgleish, H. W. P. and Häusser, M., “Simultaneous all-optical manipulation and recording of neural circuit activity with cellular resolution in vivo,” Nat. Methods 12(2), 140–146 (2015)
work page 2015
-
[5]
Yang, W., Carrillo-Reid, L., Bando, Y., Peterka, D. S. and Yuste, R., “Simultaneous two-photon imaging and two-photon optogenetics of cortical circuits in three dimensions,” eLife 7, K. Svoboda, Ed., e32671 (2018)
work page 2018
-
[6]
Precise multimodal optical control of neural ensemble activity,
Mardinly, A. R., Oldenburg, I. A., Pégard, N. C., Sridharan, S., Lyall, E. H., Chesnov, K., Brohawn, S. G., Waller, L. and Adesnik, H., “Precise multimodal optical control of neural ensemble activity,” Nat. Neurosci. 21(6), 881–893 (2018)
work page 2018
-
[7]
Single-neuron perturbations reveal feature-specific competition in V1,
Chettih, S. N. and Harvey, C. D., “Single-neuron perturbations reveal feature-specific competition in V1,” Nature 567(7748), 334–340 (2019)
work page 2019
-
[8]
Whole-brain functional imaging at cellular resolution using light-sheet microscopy,
Ahrens, M. B., Orger, M. B., Robson, D. N., Li, J. M. and Keller, P. J., “Whole-brain functional imaging at cellular resolution using light-sheet microscopy,” Nature Methods 10(5), 413–420 (2013)
work page 2013
Show all 16 references
-
[9]
Real-time generation of fully optimized holograms for optical trapping applications,
Persson, M., Engström, D. and Goksör, M., “Real-time generation of fully optimized holograms for optical trapping applications,” Proceedings of SPIE 8097, 80971H-80971H – 9 (2011)
2011
-
[10]
Improvement of the switching frequency of a liquid-crystal spatial light modulator with optimal cell gap,
Peng, Z., Liu, Y., Yao, L., Cao, Z., Mu, Q., Hu, L. and Xuan, L., “Improvement of the switching frequency of a liquid-crystal spatial light modulator with optimal cell gap,” Opt. Lett., OL 36(18), 3608–3610 (2011)
2011
-
[11]
General algorithm to optimize the diffraction efficiency of a phase-type spatial light modulator,
Cibula, M. A. and McIntyre, D. H., “General algorithm to optimize the diffraction efficiency of a phase-type spatial light modulator,” Opt. Lett., OL 38(15), 2767–2769 (2013)
2013
-
[12]
Simultaneous Multi-plane Imaging of Neural Circuits,
Yang, W., Miller, J. K., Carrillo-Reid, L., Pnevmatikakis, E., Paninski, L., Yuste, R. and Peterka, D. S., “Simultaneous Multi-plane Imaging of Neural Circuits,” Neuron 89(2), 269–284 (2016)
2016
-
[13]
Reducing the effect of pixel crosstalk in phase only spatial light modulators,
Persson, M., Engström, D. and Goksör, M., “Reducing the effect of pixel crosstalk in phase only spatial light modulators,” Opt. Express, OE 20(20), 22334–22343 (2012)
2012
-
[15]
Ultrawidefield microscope for high-speed fluorescence imaging and targeted optogenetic stimulation,
Werley, C. A., Chien, M.-P. and Cohen, A. E., “Ultrawidefield microscope for high-speed fluorescence imaging and targeted optogenetic stimulation,” Biomed Opt Express 8(12), 5794–5813 (2017)
2017
-
[16]
Speeding up liquid crystal SLMs using overdrive with phase change reduction,
Thalhammer, G., Bowman, R. W., Love, G. D., Padgett, M. J. and Ritsch-Marte, M., “Speeding up liquid crystal SLMs using overdrive with phase change reduction,” Opt Express 21(2), 1779–1797 (2013)
2013
-
[17]
Cortical layer–specific critical dynamics triggering perception,
Marshel, J. H., Kim, Y. S., Machado, T. A., Quirin, S., Benson, B., Kadmon, J., Raja, C., Chibukhchyan, A., Ramakrishnan, C., Inoue, M., Shane, J. C., McKnight, D. J., Yoshizawa, S., Kato, H. E., Ganguli, S. and Deisseroth, K., “Cortical layer–specific critical dynamics trigge...
2019
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.