¶ 1 Defendant, Delmon Edric Singley, appeals his judgment of conviction entered on a jury verdict finding him guilty of attempted second degree murder, first degree assault, attempted aggravated robbery, and felony menacing. We affirm.
I. Background
¶ 2 In the late evening of May 11, 2010, the victim, J.A.C., was commuting home from work when two men, both carrying handguns, confronted him. When J.A.C. shouted for help, one of the men opened fire, shooting him three times, fracturing his pelvic bone, and causing permanent scarring.
¶ 3 At the hospital, J.A.C. told officers that two black men in dark clothing had robbed and shot him. He described the shooter as being in his twenties with a medium-length Afro and wearing a dark-colored, unzipped, hooded sweatshirt, black shoes, and a mask.
¶ 4 Throughout the evening, officers canvassed the surrounding neighborhoods for two black men wearing dark clothing. Eventually, an officer noticed Singley, who at the time was forty-six years old and bald, and another man, wearing dark clothing and walking down the street carrying a flat-screen television and a laptop computer. When the officer confronted them, a woman ran from her home saying that the men had robbed her. Although Singley and the other man dropped the television and fled the scene, another officer was later able to arrest Singley. A search incident to arrest uncovered a mask in Singley's back pocket. During *745an interview with the arresting officer, Singley admitted to knowing about the shooting but denied any involvement in it.
¶ 5 About a week and a half after the shooting, officers presented J.A.C. with two six-photograph lineups in an attempt to identify both the shooter and his accomplice. Officers suspected Singley of being the shooter and built one lineup around a photograph of him. Within forty-five seconds, J.A.C. identified Singley as the shooter; however, he was unable to identify the accomplice among the six photographs in the second lineup. J.A.C. also identified the mask found on Singley as the one that the shooter had been wearing.
¶ 6 The prosecution's case relied heavily on J.A.C.'s prior out-of-court identification of Singley. Although J.A.C. testified to identifying Singley in the photographic lineup, he was unable to identify Singley in court. Instead, J.A.C. identified Singley from a photograph taken during his arrest on the evening of the shooting. In it, Singley is handcuffed and standing next to a police cruiser.
¶ 7 The jury convicted Singley as charged, and the trial court sentenced him to thirty-two years in the custody of the Department of Corrections for the second degree attempted murder, thirty-two years for the assault, and sixteen years for the aggravated robbery.
II. Photographic Lineup Identification
¶ 8 Singley contends that the trial court violated his right to due process and a fair trial when it declined to suppress the allegedly impermissibly suggestive and unreliable out-of-court identification, as well as the subsequent in-court identification. Although we conclude that the trial court erred when it found that the lineup was not impermissibly suggestive, we conclude that, under the totality of the circumstances, J.A.C.'s identification of Singley was nonetheless reliable and any error by the trial court in determining reliability was harmless beyond a reasonable doubt. Finally, we conclude that any error in admitting the in-court identification was not plain.
A. Standard of Review
¶ 9 The constitutionality of a pretrial identification procedure is a mixed question of law and fact. People v. Wilford, 111 P.3d 512, 514 (Colo.App.2004). While we give deference to the trial court's findings of fact, we may give weight to those facts differently and thus reach a different conclusion. People v. Hogan, 114 P.3d 42, 49 (Colo.App.2004).
¶ 10 Singley objected to the out-of-court identification; therefore, we review for nonconstitutional harmless error. Hagos v. People, 2012 CO 63, ¶ 12, 288 P.3d 116, 119. Under this standard, we reverse if there is a reasonable possibility that the erroneous admission of the identification contributed to the conviction. Id.
¶ 11 Singley did not object to J.A.C.'s in-court identification of him; therefore, we review for plain error. People v. Miller, 113 P.3d 743, 749 (Colo.2005). Plain error addresses error that is both "obvious and substantial." People v. Stewart, 55 P.3d 107, 120 (Colo.2002) (internal quotation marks omitted). It must be seriously prejudicial and " 'so undermine [ ] the fundamental fairness of the trial itself as to cast serious doubt on the reliability of the judgment of conviction.' " Hagos, ¶ 14, 288 P.3d at 120 (quoting People v. Sepulveda, 65 P.3d 1002, 1006 (Colo.2003) ).
¶ 12 Plain errors must also be "so clear-cut, [and] so obvious, that a trial judge should be able to avoid [them] without benefit of objection." People v. Pollard, 2013 COA 31M, ¶ 39, 307 P.3d 1124, 1133.
B. Applicable Law
¶ 13 The supreme court has outlined a two-part test for determining the admissibility of out-of-court photographic identifications. Bernal v. People, 44 P.3d 184, 190-91 (Colo.2002).
¶ 14 First, the defendant has the burden of demonstrating that the array was impermissibly suggestive. People v. Borghesi, 66 P.3d 93, 103 (Colo.2003). "[T]he principal question is whether the picture of the accused, which matches descriptions given by *746the witness, so stood out from all of the other photographs as to 'suggest to an identifying witness that [that person] was more likely to be the culprit.' " Bernal, 44 P.3d at 191 (quoting Jarrett v. Headley, 802 F.2d 34, 41 (2d Cir.1986) ) (some internal quotation marks omitted). Relevant factors in making this determination include "the size of the array, the manner of its presentation by the officers, and the details of the photographs themselves." Id. at 103-04 (internal quotation marks omitted). A photographic array need not contain "exact replicas" of the defendant's picture; rather, all that is required is that the "photos are matched by race, approximate age, facial hair, and a number of other characteristics." Bernal, 44 P.3d at 191-92 (internal quotation marks omitted). If a defendant cannot meet this burden and the identification is admissible, then no further inquiry is required. Id.
¶ 15 Second, if the defendant has met this burden, the state must show that the identification was nevertheless reliable under the totality of the circumstances. Id. Factors to consider include, among others, (1) the opportunity of the witness to view the criminal at the time of the crime; (2) the witness's degree of attention; (3) the accuracy of the witness's prior description of the criminal; (4) the level of certainty demonstrated by the witness at the confrontation; and (5) the length of time between the crime and the confrontation. Id. at 192. "Against these factors, courts must weigh the corrupting effect of the suggestive identification." Borghesi, 66 P.3d at 104.
C. Analysis
1. J.A.C.'s Out-of-Court Identification
¶ 16 Singley does not challenge the suggestiveness of the photographic lineup on its face; rather, he argues that the manner of presentation was impermissibly suggestive because, (1) before showing J.A.C. the array, officers told him that they had apprehended a suspect matching J.A.C.'s description on the evening of the shooting in the surrounding neighborhood; and (2) Singley, as well as the other men in the photographic lineup, did not match J.A.C.'s initial description of the shooter.
¶ 17 Because Bernal requires that the picture of the accused in any photographic lineup match the initial description given by the witness, we conclude that the trial court erred when it found that Singley did not satisfy the first prong of the Bernal test. Nevertheless, we conclude that, under the totality of the circumstances, J.A.C.'s identification of Singley was reliable.
¶ 18 The "principal question" in Bernal 's first step "is whether the picture of the accused, which matches descriptions given by the witness, so stood out from all of the other photographs." Bernal, 44 P.3d at 191 (emphasis added). We read Bernal to require that, where officers present an eyewitness with a photographic lineup, the picture of the accused must match the initial description given by the eyewitness. Id. ; see also State v. Trammell, 278 Kan. 265, 92 P.3d 1101, 1109 (Kan.2004) ("A photographic lineup is impermissibly suggestive if the photographs do not depict individuals who generally fit within the witness' description...."); but see Parsley v. State, 557 N.E.2d 1331, 1334 (Ind.1990) ("[W]hen the individuals in a lineup resemble each other the fact that the witness' initial description was somewhat different does not suggest to the witness which of the people in the lineup he should identify.").
¶ 19 Bernal 's directive that photographs of the accused in a lineup "match[ ] descriptions given by the witness" is not idle language. Bernal, 44 P.3d at 191.1 It comports with Bernal 's central purpose to ensure reliable eyewitness identifications by preventing the police from swaying a witness's memory of an assailant by implicitly suggesting an alternative description. Id. at 192 ("Reliability is the linchpin in determining the admissibility of identification testimony."). Where the police present an eyewitness *747with a photographic lineup that does not match the initial description, it may suggest that their belief as to the identity of the assailant is superior to that of the eyewitness. Such a lineup is impermissibly suggestive and violates Bernal because it suggests an alternative description of the assailant.
¶ 20 When an eyewitness describes a suspect one way, and the police present that eyewitness with a photographic lineup of suspects that do not match the description, a reasonable witness would likely decline to identify any suspect in the lineup. Where, as here, an eyewitness identifies a suspect who differs significantly from his or her initial description, a court may conclude that the photographic lineup swayed the eyewitness's recollection of the assailant, and was therefore impermissibly suggestive.
¶ 21 Therefore, before a trial court considers whether the photographs in a lineup are "matched by race, approximate age, facial hair, and ... other characteristics," it must first determine whether the photograph of the accused matches the initial description given by the eyewitness. Id. at 191-92.
¶ 22 Applying these principles, we conclude that the trial court erred when it found that Singley did not make a prima facie showing that the photographic lineup was impermissibly suggestive. Immediately after the shooting, J.A.C. told officers that the shooter was in his twenties with a medium-length Afro. Several days later, the police presented J.A.C. with a photographic lineup built around Singley, which showed six bald men, all of whom appear to be of the same general age as Singley, who was forty-six. By presenting J.A.C. with a photographic lineup of suspects that did not match his initial description of the assailant, the police implicitly proposed an alternative identification of J.A.C.'s assailant, which may have forced him to rethink and alter his initial description.
¶ 23 We recognize that the six men in the photographic lineup were all black and had facial hair, just as J.A.C. initially described. However, because the men in the lineup were twice as old as J.A.C. described and bald, we conclude that the photographs were not sufficiently similar to J.A.C.'s initial description to satisfy Bernal 's first prong.
¶ 24 Therefore, although the trial court properly concluded that Bernal 's first step requires it to look at the photographic lineup and determine whether, on its face, it is impermissibly suggestive, it erred when it did not expressly consider whether the suspects in the photographic lineup matched J.A.C.'s initial description as required by Bernal . Because the photograph of Singley did not match J.A.C.'s initial description, we conclude that the court erred when it found that Singley did not make a prima facie showing that the photographic lineup was impermissibly suggestive.
¶ 25 Having concluded that the photographic lineup was impermissibly suggestive on Bernal 's first prong, we must consider whether the identification was nevertheless reliable under the totality of the circumstances. We conclude that it was.
¶ 26 In determining whether the court erroneously denied a defendant's motion to suppress, we confine our review to the testimony developed at the suppression hearing; however, if the parties raise harmless error or plain error, we may consider the entire record. See People v. Martinez, 2015 COA 37, ¶ 15 n.5, 378 P.3d 761, 766 n.5 ; see also People v. Bowers, 801 P.2d 511, 518 (Colo.1990) (in reviewing ruling on motion in limine, a reviewing court considers initially only evidence presented at motion hearing, but if a party asserts that any error was plain or harmless, appellate court may also consider evidence presented at trial); cf. Moody v. People, 159 P.3d 611, 614 (Colo.2007) (holding that in reviewing a Fourth Amendment suppression hearing, a reviewing court focuses only on the suppression hearing record).
¶ 27 We conclude that the evidence presented at the suppression hearing supports a finding of reliability.2 J.A.C. showed a high level of certainty when identifying Singley in *748the lineup. When presented with the photographic lineup, it took J.A.C. only forty-five seconds to identify Singley without hesitation or doubt. Although the officer who presented J.A.C. with the photographic lineup told him that they had apprehended a suspect in the surrounding area on the night of the shooting, J.A.C.'s restraint with regard to the second lineup suggests that he understood the officer's admonition that the lineups may or may not have included the suspect, bolstering the certainty of his identification. See People v. Walford, 716 P.2d 137, 140 (Colo.App.1985) ("An otherwise properly conducted lineup is not constitutionally infirm where a witness knows only that a suspect has been arrested and has been included in the lineup.").
¶ 28 Even if we were to conclude that the trial court erred in concluding that J.A.C.'s out-of-court identification of Singley was reliable, we further conclude that any error was harmless beyond a reasonable doubt. See Hagos, ¶ 11.
¶ 29 In reaching this conclusion, we may consider evidence presented at the trial, as well as at the suppression hearing, since the People contend that any trial court errors were constitutionally harmless beyond a reasonable doubt. See Martinez, ¶ 15 n.5.
¶ 30 At trial, J.A.C. testified to his ample opportunity to view Singley at the time of the crime. He testified that Singley was standing only a few feet from him and, although it was nighttime, he was able to get a "somewhat" good look at the shooter because they were standing below a streetlight. He further testified that his attention was focused on the shooter, who was pointing a gun at him.
¶ 31 Also, officers arrested Singley the night of the shooting in the surrounding area. A search incident to arrest revealed that Singley was carrying a mask, which J.A.C. identified at trial as the one the shooter wore. Further, although Singley denied involvement in the shooting during an initial police interview after his arrest, Singley admitted he had knowledge of it.
¶ 32 On the other hand, we recognize that several facts caution against a finding of reliability. First, as discussed above, J.A.C.'s initial description of the shooter was different from Singley's appearance in age and hair length. Further, J.A.C. identified the shooter as wearing an unzipped hooded sweatshirt and black shoes; however, when officers arrested Singley, he was wearing a hooded sweatshirt without a zipper and white tennis shoes.
¶ 33 Second, J.A.C. testified that the shooter was wearing a mask during the holdup, which presumably hampered his ability to identify him, and J.A.C. was unable to identify Singley in court. However, while the Bernal factor of the accuracy of the witness's initial description weighs against a finding of reliability, the accuracy of J.A.C.'s initial description is more probative of weight than admissibility. Furthermore, the other factors militate towards the conclusion that J.A.C.'s identification was reliable. Combined with the fact that J.A.C. identified Singley's mask as the one the shooter wore, we conclude that J.A.C.'s out-of-court identification of Singley was reliable. In any event, considering the extensive evidence of Singley's guilt, we conclude that any error by the trial court in implicitly concluding that J.A.C.'s out-of-court identification was reliable under the totality of the was harmless beyond a reasonable doubt.
¶ 34 However, while the Bernal factor of the accuracy of the witness's initial description weighs against a finding of reliability, the accuracy of J.A.C.'s initial description is more probative of weight than admissibility. Furthermore, the other factors militate towards the conclusion that J.A.C.'s identification was reliable. Combined with the fact that J.A.C. identified Singley's mask as the one the shooter wore, we conclude that, J.A.C.'s out-of-court identification of Singley was reliable. In any event, considering the extensive evidence of Singley's guilt, we conclude that any error by the trial court in implicitly concluding that J.A.C.'s out-of-court identification was reliable under the totality of the circumstances was harmless beyond a reasonable doubt.
2. In-Court Identification
¶ 35 We further conclude that, given the reliability of the out-of-court identification, *749J.A.C.'s subsequent in-court identification of a photograph of Singley did not constitute plain error.
¶ 36 We recognize that "one-on-one show-ups," like the photograph of Singley presented to J.A.C. at trial, are disfavored "because they tend to be unnecessarily suggestive." People v. Hardiway, 874 P.2d 425, 428 (Colo.App.1993). That is especially the case here, where the photograph of Singley showed him handcuffed and standing next to a police cruiser.
¶ 37 However, even if we assume the trial court erred in permitting the photograph identification, any error does not cast serious doubt on the reliability of the judgment of conviction. Hagos, ¶ 19, 288 P.3d at 120. The jury was able to assess J.A.C.'s credibility both during his identification of Singley in the photograph, as well as during his failure to recognize Singley in court. Any prejudice resulting from the suggestive photograph was therefore offset by J.A.C.'s inability to identify Singley in court.
¶ 38 Therefore, we conclude that any error in allowing J.A.C.'s in-court identification of Singley was not plain.
III. Eyewitness Identification Instructions
¶ 39 Next, Singley contends that the trial court abused its discretion when it refused to give four proposed jury instructions on the reliability of eyewitness identification testimony. We disagree.
A. Standard of Review and Applicable Law
¶ 40 "We review jury instructions de novo to determine whether the instructions as a whole accurately inform the jury of the governing law." People v. Theus-Roberts, 2015 COA 32, ¶ 18, 378 P.3d 750. We review a trial court's decision whether to give a particular jury instruction for an abuse of discretion. Id. ; People v. Renfro, 117 P.3d 43, 48 (Colo.App.2004).
¶ 41 The supreme court has consistently held that a trial court does not abuse its discretion by refusing to give jury instructions warning of the unreliability of eyewitness identification testimony so long as it gives the pattern jury instructions on credibility and assessment of evidence. Campbell v. People, 814 P.2d 1, 5 n.8 (Colo.1991), abrogated on other grounds by People v. Shreck, 22 P.3d 68 (Colo.2001) ; People v. Fuller, 791 P.2d 702, 707 (Colo.1990) ; People v. Lopez, 182 Colo. 152, 156, 511 P.2d 889, 891 (1973) ; see also Theus-Roberts, ¶ 21, 378 P.3d at 756.
B. Analysis
¶ 42 At trial, Singley introduced expert testimony questioning the reliability of eyewitness identifications. His counsel later proffered four jury instructions on the reliability of eyewitness identifications. The trial court denied the proffered instructions, concluding, among other reasons, that they improperly highlighted the testimony of one expert witness. However, the court gave the jury a pattern witness credibility instruction, accurately informing it of the applicable law. Therefore, the trial court did not abuse its discretion when it refused to give Singley's four additional instructions.
¶ 43 We recognize that new scientific research has undermined the confidence in certain types of eyewitness identifications. Theus-Roberts, ¶¶ 42-44, 378 P.3d at 758-59 (Berger, J., specially concurring) (compiling scientific studies). This is particularly so where, as here, the eyewitness provides a cross-racial identification. Id. at ¶ 44, 378 P.3d at 759 (Studies suggest that " 'eyewitnesses are superior at identifying persons of their own race and have difficulty identifying members of another race.' ") (quoting State v. Romero, 191 N.J. 59, 922 A.2d 693, 698 (2007) ). In response to such research, eight federal circuits, including the Tenth, and seventeen states now require special jury instructions regarding eyewitness identification. Id. at ¶ 43, 378 P.3d at 758-59. We also recognize that a significant amount of time has passed since the supreme court last addressed the issue. Id. at ¶ 47, 378 P.3d at 760. However, we remain bound by the supreme court's decisions in Campbell, 791 P.2d at 707 ; Fuller, 814 P.2d at 5 n.8 ; and Lopez, 182 Colo. at 156, 511 P.2d at 891, and, based on those decisions, reaffirm that a trial court does not abuse its discretion so long as *750it gives an adequate general credibility instruction.
¶ 44 Therefore, we conclude that the trial court did not abuse its discretion when it refused to give Singley's four proposed jury instructions on the reliability of eyewitness identification testimony.
IV. Police Chief's Subpoena
¶ 45 Finally, Singley contends that the trial court abused its discretion and violated his right to present a complete defense when it quashed his subpoena of the Aurora police chief. Specifically, he asserts that the court improperly precluded the police chief's testimony regarding his assistance in helping J.A.C. obtain a U-Visa, which allowed him to reside and work legally in the United States. We disagree.
A. Standard of Review and Applicable Law
¶ 46 We review a trial court's evidentiary rulings for an abuse of discretion. People v. Melillo, 25 P.3d 769, 773 (Colo.2001). An abuse of discretion occurs when a trial court's ruling is manifestly arbitrary, unreasonable, or unfair. Id.
¶ 47 To be admissible, evidence must be relevant. CRE 402. Evidence is relevant when it has "any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence." CRE 401.
¶ 48 A trial court may exclude logically relevant evidence "if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence." CRE 403. While CRE 403 favors the admission of evidence, "the rule is an important tool to exclude matters of scant or cumulative probative force." Yusem v. People, 210 P.3d 458, 467 (Colo.2009) (internal quotation marks and alterations omitted).
¶ 49 Finally, in cases where a trial court excludes evidence, CRE 103(2) requires the objecting party to make an offer of proof, making known to the court the substance of the evidence.
B. Analysis
¶ 50 In exchange for his cooperation during the shooting investigation, members of the Aurora police department assisted J.A.C. in applying for, and obtaining, a U-Visa, which permitted him to legally reside and work in the United States.
¶ 51 At trial, Singley's counsel impeached J.A.C. with evidence of the U-Visa to establish his motive for testifying and bias. J.A.C. testified that he received the U-Visa as a result of his cooperation with the police, and that before obtaining the U-Visa he was not permitted to legally reside and work in the United States. He admitted that the U-Visa would permit him, as well as his wife and two children, to remain in the United States. He further admitted that officers from the Aurora police department had helped him obtain the U-Visa and that it was his understanding that the visa depended on his continued cooperation in the case.
¶ 52 Singley's counsel also questioned the officer who helped J.A.C. with the U-Visa application. At one point, that officer testified that it was his understanding that the U-Visa could not lead to a green card, which would permit J.A.C. to reside and work permanently within the United States.
¶ 53 Later, Singley's counsel subpoenaed the Aurora police chief to appear at trial. He argued that, because the police chief's signature appeared on the U-Visa application, he was likely to provide further testimony establishing J.A.C.'s motive to testify and bias. In his offer of proof, Singley's counsel asserted that the police chief's testimony was necessary to establish whether the Aurora police department could revoke J.A.C.'s U-Visa, or whether it could eventually lead to him receiving permanent immigration status to reside lawfully in the United States.
¶ 54 Although neither the prosecutor nor the police chief objected to the proposed testimony, the trial court, sua sponte, quashed the subpoena, ruling that the probative value of the police chief's testimony was outweighed by considerations of waste of time and the needless presentation of cumulative evidence.
¶ 55 We find no abuse of discretion. Evidence pertaining to the U-Visa was relevant *751only to establish J.A.C.'s motive and bias, which Singley's counsel had ample opportunity to question. On cross-examination, J.A.C. admitted that officers helped him obtain the U-Visa, and admitted to benefitting from it.
¶ 56 In his offer of proof, Singley's counsel argued that the police chief's testimony was necessary to elaborate on the police department's power to revoke the U-Visa and whether the visa could lead to a green card. However, although J.A.C.'s knowledge of whether the police department could revoke his U-Visa may have been relevant to his motive to testify and bias, nowhere in his offer of proof did Singley's counsel assert that the police chief told J.A.C. that he could revoke the visa. Therefore, as the trial court properly found, whether the police department could in fact revoke J.A.C.'s U-Visa was irrelevant with regard to his motive and bias in testifying.
¶ 57 Consequently, we conclude that the trial court did not abuse its discretion when it quashed Singley's subpoena of the Aurora police chief.
V. Conclusion
¶ 58 The judgment is affirmed.
JUDGE HAWTHORNE and JUDGE BERGER concur.
The supreme court cited Jarrett v. Headley, 802 F.2d 34, 41 (2d Cir.1986), for the requirement that the "picture of the accused" in a photographic lineup "match[ ] descriptions given by the witness." Bernal v. People, 44 P.3d 184, 191 (Colo.2002). However, neither Jarrett nor Bernal explains the origin of that requirement or discusses its rationale.
Even though the trial court did not expressly address the reliability prong of the Bernal test, it implicitly determined that J.A.C.'s identification of Singley was reliable, since it allowed evidence of J.A.C.'s identification to be presented to the jury.