Nodes/ComfyUI-DazzleKSampler/Dazzle TauSampler (DazzleNodes)
ComfyUI Node

Dazzle TauSampler (DazzleNodes)

A ComfyUI node in RES4LYF/samplers with 21 inputs and 3 outputs.

By DazzleNodes·Created 5 months ago·Updated 2 months ago· 3
Dazzle TauSampler (DazzleNodes)
  • model
  • positive
  • negative
  • latent_image
  • sigmas
  • guides
  • options
  • output
  • denoised
  • options
eta0.50
sampler_nametau/res_2m
schedulerbeta57
steps30
steps_to_run-1
denoise1.00
cfg5.50
seed0
latent_roleauto
sampler_modestandard
bongmathtrue
tau_versiontau2
tau_strength0.50
tau_modesoft
CategoryRES4LYF/samplers

Inputs (21)

NameTypeDefaultDescription
etaFLOAT0.50-100–100Noise amount added then removed at each step (for SDE/ancestral variants).
sampler_nameCOMBOtau/res_2m6 options: tau/res_2m, tau/res_2s, tau/dpmpp_2m, tau/dpmpp_2m_sde, tau/dpmpp_2s, tau/dpmpp_3m
schedulerCOMBObeta5710 options: simple, sgm_uniform, karras, exponential, ddim_uniform, beta, +4
stepsINT301–10000
steps_to_runINT-1-1–10000
denoiseFLOAT1.00-10000–10000
cfgFLOAT5.50-100–100
seedINT0-2–18446744073709550000Noise seed. -2 is the legacy magic value for latent-as-noise dispatch under latent_role=auto (deprecated -- set latent_role explicitly for that). Otherwise the seed drives both initial and per-step noise.
latent_roleCOMBOautoHow to interpret the input latent dict. See the DazzleKSampler tooltip for full details. 'auto' is the right default for almost all workflows. Note: latent_role controls only the initial noise tensor at sigma_max -- the seed still drives per-step ancestral/SDE noise injection regardless of role. Set eta=0 for output fully determined by upstream noise.
sampler_modeCOMBOstandard3 options: unsample, standard, resample
bongmathBOOLEANtrue
tau_versionCOMBOtau2tau1 = raw complement (noise vs coherence). tau2 = B+C structural filtering (alignment + temporal). tau3 = seed-variant perturbation via ODE chaos sensitivity (same seed, slight structural variations). tau4 = spectral per-bin complement: injects denoised's phase at frequency bands the denoiser isn't acting on this step (genuinely orthogonal via DFT basis).
tau_strengthFLOAT0.500–1Tau complement strength (0-1 range, internally mapped to effective 0-0.12). 0 = standard sampling.
tau_modeCOMBOsofthard = fixed strength every step. soft = more complement at low noise (detail phase). cosine = smooth ramp.
modeloptMODEL
positiveoptCONDITIONING
negativeoptCONDITIONING
latent_imageoptLATENT
sigmasoptSIGMAS
guidesoptGUIDES
optionsoptOPTIONS

Outputs (3)

NameTypeDescription
outputLATENT
denoisedLATENT
optionsOPTIONS